[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-arize-arize-experiment":3,"mdc--wltpu7-key":44,"related-repo-arize-arize-experiment":7063,"related-org-arize-arize-experiment":7165},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":39,"sourceUrl":42,"mdContent":43},"arize-experiment","run and analyze Arize experiments","Creates, runs, and analyzes Arize experiments for evaluating and comparing model performance. Covers experiment CRUD, exporting runs, comparing results, and evaluation workflows using the ax CLI. Use when the user mentions create experiment, run experiment, compare models, model performance, evaluate AI, experiment results, benchmark, A\u002FB test models, or measure accuracy.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"arize","Arize AI","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Farize.jpg","Arize-ai",[13,17,20],{"name":14,"slug":15,"type":16},"LLM","llm","tag",{"name":18,"slug":19,"type":16},"Evals","evals",{"name":21,"slug":22,"type":16},"Analytics","analytics",38,"https:\u002F\u002Fgithub.com\u002FArize-ai\u002Farize-skills","2026-07-31T05:53:44.725539",null,5,[29,30,31,8,32,33,34,35,36,37,38],"agent-skills","ai-agents","ai-observability","claude-code","codex","cursor","datasets","experiments","llmops","tracing",{"repoUrl":24,"stars":23,"forks":27,"topics":40,"description":41},[29,30,31,8,32,33,34,35,36,37,38],"Agent skills for Arize — datasets, experiments, and traces via the ax CLI","https:\u002F\u002Fgithub.com\u002FArize-ai\u002Farize-skills\u002Ftree\u002FHEAD\u002Fskills\u002Farize-experiment","---\nname: arize-experiment\ndescription: Creates, runs, and analyzes Arize experiments for evaluating and comparing model performance. Covers experiment CRUD, exporting runs, comparing results, and evaluation workflows using the ax CLI. Use when the user mentions create experiment, run experiment, compare models, model performance, evaluate AI, experiment results, benchmark, A\u002FB test models, or measure accuracy.\nmetadata:\n  author: arize\n  version: \"1.0\"\ncompatibility: Requires the ax CLI and a configured Arize profile.\n---\n\n# Arize Experiment Skill\n\n> **`SPACE`** — All `--space` flags and the `ARIZE_SPACE` env var accept a space **name** (e.g., `my-workspace`) or a base64 space **ID** (e.g., `U3BhY2U6...`). Find yours with `ax spaces list`.\n\n## Concepts\n\n- **Experiment** = a named evaluation run against a specific dataset version, containing one run per example\n- **Experiment Run** = the result of processing one dataset example -- includes the model output, optional evaluations, and optional metadata\n- **Dataset** = a versioned collection of examples; every experiment is tied to a dataset and a specific dataset version\n- **Evaluation** = a named metric attached to a run (e.g., `correctness`, `relevance`), with optional label, score, and explanation\n\nThe typical flow: export a dataset → process each example → collect outputs and evaluations → create an experiment with the runs.\n\n## Prerequisites\n\nProceed directly with the task — run the `ax` command you need. Do NOT check versions, env vars, or profiles upfront.\n\nIf an `ax` command fails, troubleshoot based on the error:\n- `command not found` or version error → see [references\u002Fax-setup.md](references\u002Fax-setup.md)\n- `401 Unauthorized` \u002F missing API key → run `ax profiles show` to inspect the current profile. If the profile is missing or the API key is wrong, follow [references\u002Fax-profiles.md](references\u002Fax-profiles.md) to create\u002Fupdate it. If the user doesn't have their key, direct them to https:\u002F\u002Fapp.arize.com\u002Fadmin > API Keys\n- Space unknown → run `ax spaces list` to pick by name, or ask the user\n- Project unclear → ask the user, or run `ax projects list -o json --limit 100` and present as selectable options\n- **Security:** Never read `.env` files or search the filesystem for credentials. Use `ax profiles` for Arize credentials and `ax ai-integrations` for LLM provider keys. Never ask the user to paste secrets into chat. For missing credentials, see [references\u002Fax-profiles.md](references\u002Fax-profiles.md).\n- **CRITICAL — Never fabricate outputs:** When running an experiment, you MUST call the real model API specified by the user for every dataset example. Never fabricate, simulate, or hardcode model outputs, latencies, or evaluation scores. If you cannot call the API (missing SDK, missing credentials, network error), stop and tell the user what is needed before proceeding.\n\n## List Experiments: `ax experiments list`\n\nBrowse experiments, optionally filtered by dataset. Output goes to stdout.\n\n```bash\nax experiments list\nax experiments list --dataset DATASET_NAME --space SPACE --limit 20   # DATASET_NAME: name or ID (name preferred)\nax experiments list --cursor CURSOR_TOKEN\nax experiments list -o json\n```\n\n### Flags\n\n| Flag | Type | Default | Description |\n|------|------|---------|-------------|\n| `--dataset` | string | none | Filter by dataset |\n| `--name, -n` | string | none | Substring filter on experiment name |\n| `--limit, -l` | int | 15 | Max results (1-100) |\n| `--cursor` | string | none | Pagination cursor from previous response |\n| `-o, --output` | string | table | Output format: table, json, csv, parquet, or file path |\n\n## Get Experiment: `ax experiments get`\n\nQuick metadata lookup -- returns experiment name, linked dataset\u002Fversion, and timestamps.\n\n```bash\nax experiments get NAME_OR_ID\nax experiments get NAME_OR_ID -o json\nax experiments get NAME_OR_ID --dataset DATASET_NAME --space SPACE   # required when using experiment name instead of ID\n```\n\n### Flags\n\n| Flag | Type | Default | Description |\n|------|------|---------|-------------|\n| `NAME_OR_ID` | string | required | Experiment name or ID (positional) |\n| `--dataset` | string | none | Dataset name or ID (required if using experiment name instead of ID) |\n| `--space` | string | none | Space name or ID (required if using dataset name instead of ID) |\n| `-o, --output` | string | table | Output format |\n\n### Response fields\n\n| Field | Type | Description |\n|-------|------|-------------|\n| `id` | string | Experiment ID |\n| `name` | string | Experiment name |\n| `dataset_id` | string | Linked dataset ID |\n| `dataset_version_id` | string | Specific dataset version used |\n| `experiment_traces_project_id` | string | Project where experiment traces are stored |\n| `created_at` | datetime | When the experiment was created |\n| `updated_at` | datetime | Last modification time |\n\n## Export Experiment: `ax experiments export`\n\nDownload all runs to a file. By default uses the REST API; pass `--all` to use Arrow Flight for bulk transfer.\n\n```bash\n# EXPERIMENT_NAME, DATASET_NAME: name or ID (name preferred)\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE\n# -> experiment_abc123_20260305_141500\u002Fruns.json\n\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --all\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --output-dir .\u002Fresults\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq '.[0]'\n```\n\n### Flags\n\n| Flag | Type | Default | Description |\n|------|------|---------|-------------|\n| `NAME_OR_ID` | string | required | Experiment name or ID (positional) |\n| `--dataset` | string | none | Dataset name or ID (required if using experiment name instead of ID) |\n| `--space` | string | none | Space name or ID (required if using dataset name instead of ID) |\n| `--all` | bool | false | Use Arrow Flight for bulk export (see below) |\n| `--output-dir` | string | `.` | Output directory |\n| `--stdout` | bool | false | Print JSON to stdout instead of file |\n\n### REST vs Flight (`--all`)\n\n- **REST** (default): Lower friction -- no Arrow\u002FFlight dependency, standard HTTPS ports, works through any corporate proxy or firewall. Limited to 500 runs per page.\n- **Flight** (`--all`): Required for experiments with more than 500 runs. Uses gRPC+TLS on a separate host\u002Fport which some corporate networks may block. The active `ax` profile supplies the regional endpoint; see [profile setup](references\u002Fax-profiles.md).\n\n**Agent auto-escalation rule:** If a REST export returns exactly 500 runs, the result is likely truncated. Re-run with `--all` to get the full dataset.\n\nOutput is a JSON array of run objects:\n\n```json\n[\n  {\n    \"id\": \"run_001\",\n    \"example_id\": \"ex_001\",\n    \"output\": \"The answer is 4.\",\n    \"evaluations\": {\n      \"correctness\": { \"label\": \"correct\", \"score\": 1.0 },\n      \"relevance\": { \"score\": 0.95, \"explanation\": \"Directly answers the question\" }\n    },\n    \"metadata\": { \"model\": \"gpt-4o\", \"latency_ms\": 1234 }\n  }\n]\n```\n\n## Create Experiment: `ax experiments create`\n\nCreate a new experiment with runs from a data file.\n\n```bash\nax experiments create --name \"gpt-4o-baseline\" --dataset DATASET_NAME --space SPACE --file runs.json\nax experiments create --name \"claude-test\" --dataset DATASET_NAME --space SPACE --file runs.csv\n```\n\n### Flags\n\n| Flag | Type | Required | Description |\n|------|------|----------|-------------|\n| `--name, -n` | string | yes | Experiment name |\n| `--dataset` | string | yes | Dataset to run the experiment against |\n| `--space, -s` | string | no | Space name or ID (required if using dataset name instead of ID) |\n| `--file, -f` | path | yes | Data file with runs: CSV, JSON, JSONL, or Parquet |\n| `-o, --output` | string | no | Output format |\n\n### Passing data via stdin\n\nUse `--file -` to pipe data directly — no temp file needed:\n\n```bash\necho '[{\"example_id\": \"ex_001\", \"output\": \"Paris\"}]' | ax experiments create --name \"my-experiment\" --dataset DATASET_NAME --space SPACE --file -\n\n# Or with a heredoc\nax experiments create --name \"my-experiment\" --dataset DATASET_NAME --space SPACE --file - \u003C\u003C 'EOF'\n[{\"example_id\": \"ex_001\", \"output\": \"Paris\"}]\nEOF\n```\n\n### Required columns in the runs file\n\n| Column | Type | Required | Description |\n|--------|------|----------|-------------|\n| `example_id` | string | yes | The dataset example's **top-level `id`** from `ax datasets export` |\n| `output` | string | yes | The model\u002Fsystem output for this example |\n\nAdditional columns are passed through as `additionalProperties` on the run.\n\n> **`example_id` must be the Arize row id** — the top-level `id` field on each exported dataset example (`ex[\"id\"]`). Do **not** use a value nested inside the example's input fields or `additional_properties`; a wrong value fails silently or attaches the run to the wrong example. Export the dataset and inspect the top-level `id` field before creating runs.\n\n> **⚠️ Inline evaluations in the create file do NOT attach as scores.** `create` only reads `example_id` and `output`; every other column — including an `evaluations` object — is stored as a passthrough additional field, **not** as an experiment evaluation, and will **not** appear as a score in the UI. This fails silently (no error). To attach scores\u002Flabels, create the experiment first, then run `ax experiments annotate-runs`. The `evaluations` object in the schemas below is the **export (read)** shape returned once annotations exist — it is not an input to `create`.\n\n## Delete Experiment: `ax experiments delete`\n\n```bash\nax experiments delete NAME_OR_ID\nax experiments delete NAME_OR_ID --dataset DATASET_NAME --space SPACE   # required when using experiment name instead of ID\nax experiments delete NAME_OR_ID --force   # skip confirmation prompt\n```\n\n### Flags\n\n| Flag | Type | Default | Description |\n|------|------|---------|-------------|\n| `NAME_OR_ID` | string | required | Experiment name or ID (positional) |\n| `--dataset` | string | none | Dataset name or ID (required if using experiment name instead of ID) |\n| `--space` | string | none | Space name or ID (required if using dataset name instead of ID) |\n| `--force, -f` | bool | false | Skip confirmation prompt |\n\n## Annotate Runs: `ax experiments annotate-runs`\n\n**This is the required step to attach evaluation scores\u002Flabels to an experiment and make them show up in the UI.** Evaluations cannot be attached through `create`; see the warning under Create Experiment. You write them here, after the experiment exists. Upsert semantics — resubmitting the same annotation `name` for the same run overwrites the previous value. Up to 1000 runs per request; unmatched record IDs are silently ignored.\n\n```bash\nax experiments annotate-runs NAME_OR_ID --file annotations.json --dataset DATASET_NAME --space SPACE\nax experiments annotate-runs NAME_OR_ID --file annotations.csv --dataset DATASET_NAME --space SPACE\n```\n\n### Annotation file schema\n\nA JSON array; each item annotates one run:\n\n```json\n[\n  {\n    \"record_id\": \"run_001\",\n    \"values\": [\n      { \"name\": \"correctness\", \"label\": \"correct\", \"score\": 1.0 },\n      { \"name\": \"relevance\", \"score\": 0.95, \"text\": \"Directly answers the question\" }\n    ]\n  }\n]\n```\n\n| Field | Type | Required | Description |\n|-------|------|----------|-------------|\n| `record_id` | string | yes | The **experiment run ID** (the run's `id` from `ax experiments export`) — **not** the `example_id` |\n| `values` | array | yes | One or more annotation dicts, each with a `name` plus at least one of `score`, `label`, or `text` |\n| `values[].name` | string | yes | Annotation\u002Fevaluation name (e.g., `correctness`) — becomes the score column in the UI |\n| `values[].score` | number | no | Numeric score (e.g., `0.0`–`1.0`) |\n| `values[].label` | string | no | Categorical label (e.g., `correct`, `incorrect`) |\n| `values[].text` | string | no | Freeform explanation |\n\n> `record_id` keys on the **run** id, which only exists after `create`. So the order is always: `create` → `export` (to read each run's `id`) → build annotations → `annotate-runs`.\n\n### Flags\n\n| Flag | Type | Required | Description |\n|------|------|----------|-------------|\n| `NAME_OR_ID` | string | yes | Experiment name or ID (positional) |\n| `--file, -f` | path | yes | Annotation file: JSON, JSONL, CSV, or Parquet (use `-` for stdin) |\n| `--dataset` | string | yes | Dataset name or ID (required when using experiment name instead of ID) |\n| `--space` | string | no | Space name or ID |\n\n## Experiment Run Schema\n\nEach run corresponds to one dataset example. **On `create`, only `example_id` and `output` are consumed** — `evaluations` shown here is the shape `export` returns *after* you attach scores via `annotate-runs`; it is not an input to `create`.\n\n```json\n{\n  \"example_id\": \"required on create -- the dataset example's top-level id\",\n  \"output\": \"required on create -- the model\u002Fsystem output for this example\",\n  \"evaluations\": {\n    \"metric_name\": {\n      \"label\": \"optional string label (e.g., 'correct', 'incorrect')\",\n      \"score\": \"optional numeric score (e.g., 0.95)\",\n      \"explanation\": \"optional freeform text\"\n    }\n  },\n  \"metadata\": {\n    \"model\": \"gpt-4o\",\n    \"temperature\": 0.7,\n    \"latency_ms\": 1234\n  }\n}\n```\n\n### Evaluation fields\n\n| Field | Type | Required | Description |\n|-------|------|----------|-------------|\n| `label` | string | no | Categorical classification (e.g., `correct`, `incorrect`, `partial`) |\n| `score` | number | no | Numeric quality score (e.g., 0.0 - 1.0) |\n| `explanation` | string | no | Freeform reasoning for the evaluation |\n\nAt least one of `label`, `score`, or `explanation` should be present per evaluation.\n\n## Workflows\n\n### Run an experiment against a dataset\n\n1. Find or create a dataset:\n   ```bash\n   ax datasets list --space SPACE\n   ax datasets export DATASET_NAME --space SPACE --stdout | jq 'length'\n   ```\n2. Export the dataset examples:\n   ```bash\n   ax datasets export DATASET_NAME --space SPACE\n   ```\n3. Call the real model API for each example and collect outputs. Use `ax datasets export --stdout` to pipe examples directly into an inference script:\n\n   ```bash\n   ax datasets export DATASET_NAME --space SPACE --stdout | python3 infer.py > runs.json\n   ```\n\n   Write `infer.py` to read examples from stdin, call the target model, and write runs JSON to stdout. The script below is a template — first inspect the exported dataset JSON to find the correct input field name, then uncomment the provider block the user wants:\n\n   ```python\n   import json, sys, time\n\n   examples = json.load(sys.stdin)\n   runs = []\n\n   for ex in examples:\n       # Inspect the exported JSON to find the right field (e.g. \"input\", \"question\", \"prompt\")\n       user_input = ex.get(\"input\") or ex.get(\"question\") or ex.get(\"prompt\") or str(ex)\n\n       start = time.time()\n\n       # === CALL THE REAL MODEL API HERE — never fabricate or simulate ===\n       # Uncomment and adapt the provider block the user requested:\n       #\n       # OpenAI (pip install openai  — uses OPENAI_API_KEY env var):\n       #   from openai import OpenAI\n       #   resp = OpenAI().chat.completions.create(\n       #       model=\"gpt-4o\",\n       #       messages=[{\"role\": \"user\", \"content\": user_input}]\n       #   )\n       #   output_text = resp.choices[0].message.content\n       #\n       # Anthropic (pip install anthropic  — uses ANTHROPIC_API_KEY env var):\n       #   import anthropic\n       #   resp = anthropic.Anthropic().messages.create(\n       #       model=\"claude-sonnet-4-6\", max_tokens=1024,\n       #       messages=[{\"role\": \"user\", \"content\": user_input}]\n       #   )\n       #   output_text = resp.content[0].text\n       #\n       # Google Gemini (pip install google-genai  — uses GOOGLE_API_KEY env var):\n       #   from google import genai\n       #   resp = genai.Client().models.generate_content(\n       #       model=\"gemini-2.5-pro\", contents=user_input\n       #   )\n       #   output_text = resp.text\n       #\n       # Custom \u002F OpenAI-compatible proxy (pip install openai — uses CUSTOM_BASE_URL + CUSTOM_API_KEY env vars):\n       # Use this for Azure OpenAI, NVIDIA NIM, local Ollama, or any OpenAI-compatible endpoint,\n       # including a test integration proxy. Matches the `custom` provider in `ax ai-integrations create`.\n       #   import os\n       #   from openai import OpenAI\n       #   resp = OpenAI(\n       #       base_url=os.environ[\"CUSTOM_BASE_URL\"],          # e.g. https:\u002F\u002Fmy-proxy.example.com\u002Fv1\n       #       api_key=os.environ.get(\"CUSTOM_API_KEY\", \"none\"),\n       #   ).chat.completions.create(\n       #       model=os.environ.get(\"CUSTOM_MODEL\", \"default\"),\n       #       messages=[{\"role\": \"user\", \"content\": user_input}]\n       #   )\n       #   output_text = resp.choices[0].message.content\n\n       latency_ms = round((time.time() - start) * 1000)\n       runs.append({\n           \"example_id\": ex[\"id\"],\n           \"output\": output_text,\n           \"metadata\": {\"model\": \"MODEL_NAME\", \"latency_ms\": latency_ms}\n       })\n       print(f\"  {ex['id']}: {latency_ms}ms\", file=sys.stderr)\n\n   json.dump(runs, sys.stdout, indent=2)\n   ```\n\n   **Before running:** install the provider SDK (`pip install openai` \u002F `anthropic` \u002F `google-genai`) and ensure the API key is set as an environment variable in your shell. If you cannot access the API, stop and tell the user what is needed.\n\n4. Verify the runs file:\n   ```bash\n   python3 -c \"import json; runs=json.load(open('runs.json')); print(f'{len(runs)} runs'); print(json.dumps(runs[0], indent=2))\"\n   ```\n   Each run must have `example_id` (the dataset row's top-level `id`) and `output`. `metadata` is optional. **Do not put `evaluations` here** — `create` ignores them; scores are attached in steps 7–9 below.\n5. Create the experiment:\n   ```bash\n   ax experiments create --name \"gpt-4o-baseline\" --dataset DATASET_NAME --space SPACE --file runs.json\n   ```\n6. Verify: `ax experiments get \"gpt-4o-baseline\" --dataset DATASET_NAME --space SPACE`\n\n   **Attach evaluation scores (required for scores to show in the UI).** Evaluations do **not** come from the create file — you attach them with `annotate-runs`, which keys on each run's `id` (assigned at create time), so you must export first to learn those IDs.\n\n7. Export the experiment to structured data so you can read each run's `id` alongside its `example_id`. Confirm that the exported run records include both fields.\n8. Build the annotation file with structured JSON handling, keyed by `record_id` (the run `id`). Score\u002Flabel each run via an LLM-as-judge, a code check, or human review; never fabricate scores. Emit this shape:\n   ```json\n   [\n     {\n       \"record_id\": \"RUN_ID_FROM_EXPERIMENT_EXPORT\",\n       \"values\": [\n         { \"name\": \"correctness\", \"score\": 1.0, \"label\": \"correct\" }\n       ]\n     }\n   ]\n   ```\n9. Attach the scores with `ax experiments annotate-runs ... --file annotations.json`, then export or inspect the experiment to confirm the evaluations are attached.\n   The scores now render in the experiment view in the Arize UI.\n\n### Compare two experiments\n\n1. Export both experiments:\n   ```bash\n   ax experiments export \"experiment-a\" --dataset DATASET_NAME --space SPACE --stdout > a.json\n   ax experiments export \"experiment-b\" --dataset DATASET_NAME --space SPACE --stdout > b.json\n   ```\n2. Compare evaluation scores by `example_id`:\n   ```bash\n   # Average correctness score for experiment A\n   jq '[.[] | .evaluations.correctness.score] | add \u002F length' a.json\n\n   # Same for experiment B\n   jq '[.[] | .evaluations.correctness.score] | add \u002F length' b.json\n   ```\n3. Find examples where results differ:\n   ```bash\n   jq -s '.[0] as $a | .[1][] | . as $run |\n     {\n       example_id: $run.example_id,\n       b_score: $run.evaluations.correctness.score,\n       a_score: ($a[] | select(.example_id == $run.example_id) | .evaluations.correctness.score)\n     }' a.json b.json\n   ```\n4. Score distribution per evaluator (pass\u002Ffail\u002Fpartial counts):\n   ```bash\n   # Count by label for experiment A\n   jq '[.[] | .evaluations.correctness.label] | group_by(.) | map({label: .[0], count: length})' a.json\n   ```\n5. Find regressions (examples that passed in A but fail in B):\n   ```bash\n   jq -s '\n     [.[0][] | select(.evaluations.correctness.label == \"correct\")] as $passed_a |\n     [.[1][] | select(.evaluations.correctness.label != \"correct\") |\n       select(.example_id as $id | $passed_a | any(.example_id == $id))\n     ]\n   ' a.json b.json\n   ```\n\n**Statistical significance note:** Score comparisons are most reliable with ≥ 30 examples per evaluator. With fewer examples, treat the delta as directional only — a 5% difference on n=10 may be noise. Report sample size alongside scores: `jq 'length' a.json`.\n\n### Download experiment results for analysis\n\n1. `ax experiments list --dataset DATASET_NAME --space SPACE` -- find experiments\n2. `ax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE` -- download to file\n3. Parse: `jq '.[] | {example_id, score: .evaluations.correctness.score}' experiment_*\u002Fruns.json`\n\n### Pipe export to other tools\n\n```bash\n# Count runs\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq 'length'\n\n# Extract all outputs\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq '.[].output'\n\n# Get runs with low scores\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq '[.[] | select(.evaluations.correctness.score \u003C 0.5)]'\n\n# Convert to CSV\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq -r '.[] | [.example_id, .output, .evaluations.correctness.score] | @csv'\n```\n\n## Related Skills\n\n- **arize-dataset**: Create or export the dataset this experiment runs against → use `arize-dataset` first\n- **arize-prompts**: Store and version the prompt template in Prompt Hub (`ax prompts`) before or after experiments\n- **arize-prompt-optimization**: Use experiment results to improve prompts → next step is `arize-prompt-optimization`\n- **arize-trace**: Inspect individual span traces for failing experiment runs → use `arize-trace`\n- **arize-link**: Generate clickable UI links to traces from experiment runs → use `arize-link`\n\n## Troubleshooting\n\n| Problem | Solution |\n|---------|----------|\n| `ax: command not found` | See [references\u002Fax-setup.md](references\u002Fax-setup.md) |\n| `401 Unauthorized` | API key is wrong, expired, or doesn't have access to this space. Fix the profile using [references\u002Fax-profiles.md](references\u002Fax-profiles.md). |\n| `No profile found` | No profile is configured. See [references\u002Fax-profiles.md](references\u002Fax-profiles.md) to create one. |\n| `Experiment not found` | Verify experiment name with `ax experiments list --space SPACE` |\n| `Invalid runs file` | Each run must have `example_id` and `output` fields |\n| `example_id mismatch` | `example_id` must be the dataset row's **top-level `id`** from `ax datasets export` — not a value nested in the example's fields or `additional_properties`. Export the dataset and inspect the top-level `id` field. |\n| Runs created but no scores \u002F evals in the UI | Evaluations in the create file are silently ignored. Attach them with `ax experiments annotate-runs` (keyed by run `id`) after creating the experiment — see the workflow steps 7–9. |\n| `annotate-runs` reports success but nothing changes | `record_id` must be the **run `id`** (from `ax experiments export`), not the `example_id`. Unmatched record IDs are silently ignored. |\n| `No runs found` | Export returned empty -- verify experiment has runs via `ax experiments get` |\n| `Dataset not found` | The linked dataset may have been deleted; check with `ax datasets list` |\n\n## Save Credentials for Future Use\n\nSee [references\u002Fax-profiles.md](references\u002Fax-profiles.md) § Save Credentials for Future Use.\n",{"data":45,"body":49},{"name":4,"description":6,"metadata":46,"compatibility":48},{"author":8,"version":47},"1.0","Requires the ax CLI and a configured Arize profile.",{"type":50,"children":51},"root",[52,61,135,142,203,208,214,227,239,368,380,385,524,531,698,710,715,811,816,942,948,1120,1132,1145,1409,1414,1591,1604,1647,1664,1669,2117,2129,2134,2258,2263,2414,2420,2433,2619,2625,2720,2733,2788,2870,2882,2977,2982,3104,3115,3139,3235,3241,3246,3541,3809,3865,3870,4001,4007,4074,4460,4466,4585,4609,4615,4621,5835,5841,6264,6281,6287,6323,6329,6633,6639,6722,6728,7041,7047,7057],{"type":53,"tag":54,"props":55,"children":57},"element","h1",{"id":56},"arize-experiment-skill",[58],{"type":59,"value":60},"text","Arize Experiment Skill",{"type":53,"tag":62,"props":63,"children":64},"blockquote",{},[65],{"type":53,"tag":66,"props":67,"children":68},"p",{},[69,80,82,88,90,96,98,103,105,111,113,118,119,125,127,133],{"type":53,"tag":70,"props":71,"children":72},"strong",{},[73],{"type":53,"tag":74,"props":75,"children":77},"code",{"className":76},[],[78],{"type":59,"value":79},"SPACE",{"type":59,"value":81}," — All ",{"type":53,"tag":74,"props":83,"children":85},{"className":84},[],[86],{"type":59,"value":87},"--space",{"type":59,"value":89}," flags and the ",{"type":53,"tag":74,"props":91,"children":93},{"className":92},[],[94],{"type":59,"value":95},"ARIZE_SPACE",{"type":59,"value":97}," env var accept a space ",{"type":53,"tag":70,"props":99,"children":100},{},[101],{"type":59,"value":102},"name",{"type":59,"value":104}," (e.g., ",{"type":53,"tag":74,"props":106,"children":108},{"className":107},[],[109],{"type":59,"value":110},"my-workspace",{"type":59,"value":112},") or a base64 space ",{"type":53,"tag":70,"props":114,"children":115},{},[116],{"type":59,"value":117},"ID",{"type":59,"value":104},{"type":53,"tag":74,"props":120,"children":122},{"className":121},[],[123],{"type":59,"value":124},"U3BhY2U6...",{"type":59,"value":126},"). Find yours with ",{"type":53,"tag":74,"props":128,"children":130},{"className":129},[],[131],{"type":59,"value":132},"ax spaces list",{"type":59,"value":134},".",{"type":53,"tag":136,"props":137,"children":139},"h2",{"id":138},"concepts",[140],{"type":59,"value":141},"Concepts",{"type":53,"tag":143,"props":144,"children":145},"ul",{},[146,157,167,177],{"type":53,"tag":147,"props":148,"children":149},"li",{},[150,155],{"type":53,"tag":70,"props":151,"children":152},{},[153],{"type":59,"value":154},"Experiment",{"type":59,"value":156}," = a named evaluation run against a specific dataset version, containing one run per example",{"type":53,"tag":147,"props":158,"children":159},{},[160,165],{"type":53,"tag":70,"props":161,"children":162},{},[163],{"type":59,"value":164},"Experiment Run",{"type":59,"value":166}," = the result of processing one dataset example -- includes the model output, optional evaluations, and optional metadata",{"type":53,"tag":147,"props":168,"children":169},{},[170,175],{"type":53,"tag":70,"props":171,"children":172},{},[173],{"type":59,"value":174},"Dataset",{"type":59,"value":176}," = a versioned collection of examples; every experiment is tied to a dataset and a specific dataset version",{"type":53,"tag":147,"props":178,"children":179},{},[180,185,187,193,195,201],{"type":53,"tag":70,"props":181,"children":182},{},[183],{"type":59,"value":184},"Evaluation",{"type":59,"value":186}," = a named metric attached to a run (e.g., ",{"type":53,"tag":74,"props":188,"children":190},{"className":189},[],[191],{"type":59,"value":192},"correctness",{"type":59,"value":194},", ",{"type":53,"tag":74,"props":196,"children":198},{"className":197},[],[199],{"type":59,"value":200},"relevance",{"type":59,"value":202},"), with optional label, score, and explanation",{"type":53,"tag":66,"props":204,"children":205},{},[206],{"type":59,"value":207},"The typical flow: export a dataset → process each example → collect outputs and evaluations → create an experiment with the runs.",{"type":53,"tag":136,"props":209,"children":211},{"id":210},"prerequisites",[212],{"type":59,"value":213},"Prerequisites",{"type":53,"tag":66,"props":215,"children":216},{},[217,219,225],{"type":59,"value":218},"Proceed directly with the task — run the ",{"type":53,"tag":74,"props":220,"children":222},{"className":221},[],[223],{"type":59,"value":224},"ax",{"type":59,"value":226}," command you need. Do NOT check versions, env vars, or profiles upfront.",{"type":53,"tag":66,"props":228,"children":229},{},[230,232,237],{"type":59,"value":231},"If an ",{"type":53,"tag":74,"props":233,"children":235},{"className":234},[],[236],{"type":59,"value":224},{"type":59,"value":238}," command fails, troubleshoot based on the error:",{"type":53,"tag":143,"props":240,"children":241},{},[242,259,294,306,319,358],{"type":53,"tag":147,"props":243,"children":244},{},[245,251,253],{"type":53,"tag":74,"props":246,"children":248},{"className":247},[],[249],{"type":59,"value":250},"command not found",{"type":59,"value":252}," or version error → see ",{"type":53,"tag":254,"props":255,"children":257},"a",{"href":256},"references\u002Fax-setup.md",[258],{"type":59,"value":256},{"type":53,"tag":147,"props":260,"children":261},{},[262,268,270,276,278,283,285,292],{"type":53,"tag":74,"props":263,"children":265},{"className":264},[],[266],{"type":59,"value":267},"401 Unauthorized",{"type":59,"value":269}," \u002F missing API key → run ",{"type":53,"tag":74,"props":271,"children":273},{"className":272},[],[274],{"type":59,"value":275},"ax profiles show",{"type":59,"value":277}," to inspect the current profile. If the profile is missing or the API key is wrong, follow ",{"type":53,"tag":254,"props":279,"children":281},{"href":280},"references\u002Fax-profiles.md",[282],{"type":59,"value":280},{"type":59,"value":284}," to create\u002Fupdate it. If the user doesn't have their key, direct them to ",{"type":53,"tag":254,"props":286,"children":290},{"href":287,"rel":288},"https:\u002F\u002Fapp.arize.com\u002Fadmin",[289],"nofollow",[291],{"type":59,"value":287},{"type":59,"value":293}," > API Keys",{"type":53,"tag":147,"props":295,"children":296},{},[297,299,304],{"type":59,"value":298},"Space unknown → run ",{"type":53,"tag":74,"props":300,"children":302},{"className":301},[],[303],{"type":59,"value":132},{"type":59,"value":305}," to pick by name, or ask the user",{"type":53,"tag":147,"props":307,"children":308},{},[309,311,317],{"type":59,"value":310},"Project unclear → ask the user, or run ",{"type":53,"tag":74,"props":312,"children":314},{"className":313},[],[315],{"type":59,"value":316},"ax projects list -o json --limit 100",{"type":59,"value":318}," and present as selectable options",{"type":53,"tag":147,"props":320,"children":321},{},[322,327,329,335,337,343,345,351,353,357],{"type":53,"tag":70,"props":323,"children":324},{},[325],{"type":59,"value":326},"Security:",{"type":59,"value":328}," Never read ",{"type":53,"tag":74,"props":330,"children":332},{"className":331},[],[333],{"type":59,"value":334},".env",{"type":59,"value":336}," files or search the filesystem for credentials. Use ",{"type":53,"tag":74,"props":338,"children":340},{"className":339},[],[341],{"type":59,"value":342},"ax profiles",{"type":59,"value":344}," for Arize credentials and ",{"type":53,"tag":74,"props":346,"children":348},{"className":347},[],[349],{"type":59,"value":350},"ax ai-integrations",{"type":59,"value":352}," for LLM provider keys. Never ask the user to paste secrets into chat. For missing credentials, see ",{"type":53,"tag":254,"props":354,"children":355},{"href":280},[356],{"type":59,"value":280},{"type":59,"value":134},{"type":53,"tag":147,"props":359,"children":360},{},[361,366],{"type":53,"tag":70,"props":362,"children":363},{},[364],{"type":59,"value":365},"CRITICAL — Never fabricate outputs:",{"type":59,"value":367}," When running an experiment, you MUST call the real model API specified by the user for every dataset example. Never fabricate, simulate, or hardcode model outputs, latencies, or evaluation scores. If you cannot call the API (missing SDK, missing credentials, network error), stop and tell the user what is needed before proceeding.",{"type":53,"tag":136,"props":369,"children":371},{"id":370},"list-experiments-ax-experiments-list",[372,374],{"type":59,"value":373},"List Experiments: ",{"type":53,"tag":74,"props":375,"children":377},{"className":376},[],[378],{"type":59,"value":379},"ax experiments list",{"type":53,"tag":66,"props":381,"children":382},{},[383],{"type":59,"value":384},"Browse experiments, optionally filtered by dataset. Output goes to stdout.",{"type":53,"tag":386,"props":387,"children":392},"pre",{"className":388,"code":389,"language":390,"meta":391,"style":391},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","ax experiments list\nax experiments list --dataset DATASET_NAME --space SPACE --limit 20   # DATASET_NAME: name or ID (name preferred)\nax experiments list --cursor CURSOR_TOKEN\nax experiments list -o json\n","bash","",[393],{"type":53,"tag":74,"props":394,"children":395},{"__ignoreMap":391},[396,418,472,498],{"type":53,"tag":397,"props":398,"children":401},"span",{"class":399,"line":400},"line",1,[402,407,413],{"type":53,"tag":397,"props":403,"children":405},{"style":404},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[406],{"type":59,"value":224},{"type":53,"tag":397,"props":408,"children":410},{"style":409},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[411],{"type":59,"value":412}," experiments",{"type":53,"tag":397,"props":414,"children":415},{"style":409},[416],{"type":59,"value":417}," list\n",{"type":53,"tag":397,"props":419,"children":421},{"class":399,"line":420},2,[422,426,430,435,440,445,450,455,460,466],{"type":53,"tag":397,"props":423,"children":424},{"style":404},[425],{"type":59,"value":224},{"type":53,"tag":397,"props":427,"children":428},{"style":409},[429],{"type":59,"value":412},{"type":53,"tag":397,"props":431,"children":432},{"style":409},[433],{"type":59,"value":434}," list",{"type":53,"tag":397,"props":436,"children":437},{"style":409},[438],{"type":59,"value":439}," --dataset",{"type":53,"tag":397,"props":441,"children":442},{"style":409},[443],{"type":59,"value":444}," DATASET_NAME",{"type":53,"tag":397,"props":446,"children":447},{"style":409},[448],{"type":59,"value":449}," --space",{"type":53,"tag":397,"props":451,"children":452},{"style":409},[453],{"type":59,"value":454}," SPACE",{"type":53,"tag":397,"props":456,"children":457},{"style":409},[458],{"type":59,"value":459}," --limit",{"type":53,"tag":397,"props":461,"children":463},{"style":462},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[464],{"type":59,"value":465}," 20",{"type":53,"tag":397,"props":467,"children":469},{"style":468},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[470],{"type":59,"value":471},"   # DATASET_NAME: name or ID (name preferred)\n",{"type":53,"tag":397,"props":473,"children":475},{"class":399,"line":474},3,[476,480,484,488,493],{"type":53,"tag":397,"props":477,"children":478},{"style":404},[479],{"type":59,"value":224},{"type":53,"tag":397,"props":481,"children":482},{"style":409},[483],{"type":59,"value":412},{"type":53,"tag":397,"props":485,"children":486},{"style":409},[487],{"type":59,"value":434},{"type":53,"tag":397,"props":489,"children":490},{"style":409},[491],{"type":59,"value":492}," --cursor",{"type":53,"tag":397,"props":494,"children":495},{"style":409},[496],{"type":59,"value":497}," CURSOR_TOKEN\n",{"type":53,"tag":397,"props":499,"children":501},{"class":399,"line":500},4,[502,506,510,514,519],{"type":53,"tag":397,"props":503,"children":504},{"style":404},[505],{"type":59,"value":224},{"type":53,"tag":397,"props":507,"children":508},{"style":409},[509],{"type":59,"value":412},{"type":53,"tag":397,"props":511,"children":512},{"style":409},[513],{"type":59,"value":434},{"type":53,"tag":397,"props":515,"children":516},{"style":409},[517],{"type":59,"value":518}," -o",{"type":53,"tag":397,"props":520,"children":521},{"style":409},[522],{"type":59,"value":523}," json\n",{"type":53,"tag":525,"props":526,"children":528},"h3",{"id":527},"flags",[529],{"type":59,"value":530},"Flags",{"type":53,"tag":532,"props":533,"children":534},"table",{},[535,564],{"type":53,"tag":536,"props":537,"children":538},"thead",{},[539],{"type":53,"tag":540,"props":541,"children":542},"tr",{},[543,549,554,559],{"type":53,"tag":544,"props":545,"children":546},"th",{},[547],{"type":59,"value":548},"Flag",{"type":53,"tag":544,"props":550,"children":551},{},[552],{"type":59,"value":553},"Type",{"type":53,"tag":544,"props":555,"children":556},{},[557],{"type":59,"value":558},"Default",{"type":53,"tag":544,"props":560,"children":561},{},[562],{"type":59,"value":563},"Description",{"type":53,"tag":565,"props":566,"children":567},"tbody",{},[568,596,621,648,673],{"type":53,"tag":540,"props":569,"children":570},{},[571,581,586,591],{"type":53,"tag":572,"props":573,"children":574},"td",{},[575],{"type":53,"tag":74,"props":576,"children":578},{"className":577},[],[579],{"type":59,"value":580},"--dataset",{"type":53,"tag":572,"props":582,"children":583},{},[584],{"type":59,"value":585},"string",{"type":53,"tag":572,"props":587,"children":588},{},[589],{"type":59,"value":590},"none",{"type":53,"tag":572,"props":592,"children":593},{},[594],{"type":59,"value":595},"Filter by dataset",{"type":53,"tag":540,"props":597,"children":598},{},[599,608,612,616],{"type":53,"tag":572,"props":600,"children":601},{},[602],{"type":53,"tag":74,"props":603,"children":605},{"className":604},[],[606],{"type":59,"value":607},"--name, -n",{"type":53,"tag":572,"props":609,"children":610},{},[611],{"type":59,"value":585},{"type":53,"tag":572,"props":613,"children":614},{},[615],{"type":59,"value":590},{"type":53,"tag":572,"props":617,"children":618},{},[619],{"type":59,"value":620},"Substring filter on experiment name",{"type":53,"tag":540,"props":622,"children":623},{},[624,633,638,643],{"type":53,"tag":572,"props":625,"children":626},{},[627],{"type":53,"tag":74,"props":628,"children":630},{"className":629},[],[631],{"type":59,"value":632},"--limit, -l",{"type":53,"tag":572,"props":634,"children":635},{},[636],{"type":59,"value":637},"int",{"type":53,"tag":572,"props":639,"children":640},{},[641],{"type":59,"value":642},"15",{"type":53,"tag":572,"props":644,"children":645},{},[646],{"type":59,"value":647},"Max results (1-100)",{"type":53,"tag":540,"props":649,"children":650},{},[651,660,664,668],{"type":53,"tag":572,"props":652,"children":653},{},[654],{"type":53,"tag":74,"props":655,"children":657},{"className":656},[],[658],{"type":59,"value":659},"--cursor",{"type":53,"tag":572,"props":661,"children":662},{},[663],{"type":59,"value":585},{"type":53,"tag":572,"props":665,"children":666},{},[667],{"type":59,"value":590},{"type":53,"tag":572,"props":669,"children":670},{},[671],{"type":59,"value":672},"Pagination cursor from previous response",{"type":53,"tag":540,"props":674,"children":675},{},[676,685,689,693],{"type":53,"tag":572,"props":677,"children":678},{},[679],{"type":53,"tag":74,"props":680,"children":682},{"className":681},[],[683],{"type":59,"value":684},"-o, --output",{"type":53,"tag":572,"props":686,"children":687},{},[688],{"type":59,"value":585},{"type":53,"tag":572,"props":690,"children":691},{},[692],{"type":59,"value":532},{"type":53,"tag":572,"props":694,"children":695},{},[696],{"type":59,"value":697},"Output format: table, json, csv, parquet, or file path",{"type":53,"tag":136,"props":699,"children":701},{"id":700},"get-experiment-ax-experiments-get",[702,704],{"type":59,"value":703},"Get Experiment: ",{"type":53,"tag":74,"props":705,"children":707},{"className":706},[],[708],{"type":59,"value":709},"ax experiments get",{"type":53,"tag":66,"props":711,"children":712},{},[713],{"type":59,"value":714},"Quick metadata lookup -- returns experiment name, linked dataset\u002Fversion, and timestamps.",{"type":53,"tag":386,"props":716,"children":718},{"className":388,"code":717,"language":390,"meta":391,"style":391},"ax experiments get NAME_OR_ID\nax experiments get NAME_OR_ID -o json\nax experiments get NAME_OR_ID --dataset DATASET_NAME --space SPACE   # required when using experiment name instead of ID\n",[719],{"type":53,"tag":74,"props":720,"children":721},{"__ignoreMap":391},[722,743,771],{"type":53,"tag":397,"props":723,"children":724},{"class":399,"line":400},[725,729,733,738],{"type":53,"tag":397,"props":726,"children":727},{"style":404},[728],{"type":59,"value":224},{"type":53,"tag":397,"props":730,"children":731},{"style":409},[732],{"type":59,"value":412},{"type":53,"tag":397,"props":734,"children":735},{"style":409},[736],{"type":59,"value":737}," get",{"type":53,"tag":397,"props":739,"children":740},{"style":409},[741],{"type":59,"value":742}," NAME_OR_ID\n",{"type":53,"tag":397,"props":744,"children":745},{"class":399,"line":420},[746,750,754,758,763,767],{"type":53,"tag":397,"props":747,"children":748},{"style":404},[749],{"type":59,"value":224},{"type":53,"tag":397,"props":751,"children":752},{"style":409},[753],{"type":59,"value":412},{"type":53,"tag":397,"props":755,"children":756},{"style":409},[757],{"type":59,"value":737},{"type":53,"tag":397,"props":759,"children":760},{"style":409},[761],{"type":59,"value":762}," NAME_OR_ID",{"type":53,"tag":397,"props":764,"children":765},{"style":409},[766],{"type":59,"value":518},{"type":53,"tag":397,"props":768,"children":769},{"style":409},[770],{"type":59,"value":523},{"type":53,"tag":397,"props":772,"children":773},{"class":399,"line":474},[774,778,782,786,790,794,798,802,806],{"type":53,"tag":397,"props":775,"children":776},{"style":404},[777],{"type":59,"value":224},{"type":53,"tag":397,"props":779,"children":780},{"style":409},[781],{"type":59,"value":412},{"type":53,"tag":397,"props":783,"children":784},{"style":409},[785],{"type":59,"value":737},{"type":53,"tag":397,"props":787,"children":788},{"style":409},[789],{"type":59,"value":762},{"type":53,"tag":397,"props":791,"children":792},{"style":409},[793],{"type":59,"value":439},{"type":53,"tag":397,"props":795,"children":796},{"style":409},[797],{"type":59,"value":444},{"type":53,"tag":397,"props":799,"children":800},{"style":409},[801],{"type":59,"value":449},{"type":53,"tag":397,"props":803,"children":804},{"style":409},[805],{"type":59,"value":454},{"type":53,"tag":397,"props":807,"children":808},{"style":468},[809],{"type":59,"value":810},"   # required when using experiment name instead of ID\n",{"type":53,"tag":525,"props":812,"children":814},{"id":813},"flags-1",[815],{"type":59,"value":530},{"type":53,"tag":532,"props":817,"children":818},{},[819,841],{"type":53,"tag":536,"props":820,"children":821},{},[822],{"type":53,"tag":540,"props":823,"children":824},{},[825,829,833,837],{"type":53,"tag":544,"props":826,"children":827},{},[828],{"type":59,"value":548},{"type":53,"tag":544,"props":830,"children":831},{},[832],{"type":59,"value":553},{"type":53,"tag":544,"props":834,"children":835},{},[836],{"type":59,"value":558},{"type":53,"tag":544,"props":838,"children":839},{},[840],{"type":59,"value":563},{"type":53,"tag":565,"props":842,"children":843},{},[844,870,894,918],{"type":53,"tag":540,"props":845,"children":846},{},[847,856,860,865],{"type":53,"tag":572,"props":848,"children":849},{},[850],{"type":53,"tag":74,"props":851,"children":853},{"className":852},[],[854],{"type":59,"value":855},"NAME_OR_ID",{"type":53,"tag":572,"props":857,"children":858},{},[859],{"type":59,"value":585},{"type":53,"tag":572,"props":861,"children":862},{},[863],{"type":59,"value":864},"required",{"type":53,"tag":572,"props":866,"children":867},{},[868],{"type":59,"value":869},"Experiment name or ID (positional)",{"type":53,"tag":540,"props":871,"children":872},{},[873,881,885,889],{"type":53,"tag":572,"props":874,"children":875},{},[876],{"type":53,"tag":74,"props":877,"children":879},{"className":878},[],[880],{"type":59,"value":580},{"type":53,"tag":572,"props":882,"children":883},{},[884],{"type":59,"value":585},{"type":53,"tag":572,"props":886,"children":887},{},[888],{"type":59,"value":590},{"type":53,"tag":572,"props":890,"children":891},{},[892],{"type":59,"value":893},"Dataset name or ID (required if using experiment name instead of ID)",{"type":53,"tag":540,"props":895,"children":896},{},[897,905,909,913],{"type":53,"tag":572,"props":898,"children":899},{},[900],{"type":53,"tag":74,"props":901,"children":903},{"className":902},[],[904],{"type":59,"value":87},{"type":53,"tag":572,"props":906,"children":907},{},[908],{"type":59,"value":585},{"type":53,"tag":572,"props":910,"children":911},{},[912],{"type":59,"value":590},{"type":53,"tag":572,"props":914,"children":915},{},[916],{"type":59,"value":917},"Space name or ID (required if using dataset name instead of ID)",{"type":53,"tag":540,"props":919,"children":920},{},[921,929,933,937],{"type":53,"tag":572,"props":922,"children":923},{},[924],{"type":53,"tag":74,"props":925,"children":927},{"className":926},[],[928],{"type":59,"value":684},{"type":53,"tag":572,"props":930,"children":931},{},[932],{"type":59,"value":585},{"type":53,"tag":572,"props":934,"children":935},{},[936],{"type":59,"value":532},{"type":53,"tag":572,"props":938,"children":939},{},[940],{"type":59,"value":941},"Output format",{"type":53,"tag":525,"props":943,"children":945},{"id":944},"response-fields",[946],{"type":59,"value":947},"Response fields",{"type":53,"tag":532,"props":949,"children":950},{},[951,970],{"type":53,"tag":536,"props":952,"children":953},{},[954],{"type":53,"tag":540,"props":955,"children":956},{},[957,962,966],{"type":53,"tag":544,"props":958,"children":959},{},[960],{"type":59,"value":961},"Field",{"type":53,"tag":544,"props":963,"children":964},{},[965],{"type":59,"value":553},{"type":53,"tag":544,"props":967,"children":968},{},[969],{"type":59,"value":563},{"type":53,"tag":565,"props":971,"children":972},{},[973,994,1014,1035,1056,1077,1099],{"type":53,"tag":540,"props":974,"children":975},{},[976,985,989],{"type":53,"tag":572,"props":977,"children":978},{},[979],{"type":53,"tag":74,"props":980,"children":982},{"className":981},[],[983],{"type":59,"value":984},"id",{"type":53,"tag":572,"props":986,"children":987},{},[988],{"type":59,"value":585},{"type":53,"tag":572,"props":990,"children":991},{},[992],{"type":59,"value":993},"Experiment ID",{"type":53,"tag":540,"props":995,"children":996},{},[997,1005,1009],{"type":53,"tag":572,"props":998,"children":999},{},[1000],{"type":53,"tag":74,"props":1001,"children":1003},{"className":1002},[],[1004],{"type":59,"value":102},{"type":53,"tag":572,"props":1006,"children":1007},{},[1008],{"type":59,"value":585},{"type":53,"tag":572,"props":1010,"children":1011},{},[1012],{"type":59,"value":1013},"Experiment name",{"type":53,"tag":540,"props":1015,"children":1016},{},[1017,1026,1030],{"type":53,"tag":572,"props":1018,"children":1019},{},[1020],{"type":53,"tag":74,"props":1021,"children":1023},{"className":1022},[],[1024],{"type":59,"value":1025},"dataset_id",{"type":53,"tag":572,"props":1027,"children":1028},{},[1029],{"type":59,"value":585},{"type":53,"tag":572,"props":1031,"children":1032},{},[1033],{"type":59,"value":1034},"Linked dataset ID",{"type":53,"tag":540,"props":1036,"children":1037},{},[1038,1047,1051],{"type":53,"tag":572,"props":1039,"children":1040},{},[1041],{"type":53,"tag":74,"props":1042,"children":1044},{"className":1043},[],[1045],{"type":59,"value":1046},"dataset_version_id",{"type":53,"tag":572,"props":1048,"children":1049},{},[1050],{"type":59,"value":585},{"type":53,"tag":572,"props":1052,"children":1053},{},[1054],{"type":59,"value":1055},"Specific dataset version used",{"type":53,"tag":540,"props":1057,"children":1058},{},[1059,1068,1072],{"type":53,"tag":572,"props":1060,"children":1061},{},[1062],{"type":53,"tag":74,"props":1063,"children":1065},{"className":1064},[],[1066],{"type":59,"value":1067},"experiment_traces_project_id",{"type":53,"tag":572,"props":1069,"children":1070},{},[1071],{"type":59,"value":585},{"type":53,"tag":572,"props":1073,"children":1074},{},[1075],{"type":59,"value":1076},"Project where experiment traces are stored",{"type":53,"tag":540,"props":1078,"children":1079},{},[1080,1089,1094],{"type":53,"tag":572,"props":1081,"children":1082},{},[1083],{"type":53,"tag":74,"props":1084,"children":1086},{"className":1085},[],[1087],{"type":59,"value":1088},"created_at",{"type":53,"tag":572,"props":1090,"children":1091},{},[1092],{"type":59,"value":1093},"datetime",{"type":53,"tag":572,"props":1095,"children":1096},{},[1097],{"type":59,"value":1098},"When the experiment was created",{"type":53,"tag":540,"props":1100,"children":1101},{},[1102,1111,1115],{"type":53,"tag":572,"props":1103,"children":1104},{},[1105],{"type":53,"tag":74,"props":1106,"children":1108},{"className":1107},[],[1109],{"type":59,"value":1110},"updated_at",{"type":53,"tag":572,"props":1112,"children":1113},{},[1114],{"type":59,"value":1093},{"type":53,"tag":572,"props":1116,"children":1117},{},[1118],{"type":59,"value":1119},"Last modification time",{"type":53,"tag":136,"props":1121,"children":1123},{"id":1122},"export-experiment-ax-experiments-export",[1124,1126],{"type":59,"value":1125},"Export Experiment: ",{"type":53,"tag":74,"props":1127,"children":1129},{"className":1128},[],[1130],{"type":59,"value":1131},"ax experiments export",{"type":53,"tag":66,"props":1133,"children":1134},{},[1135,1137,1143],{"type":59,"value":1136},"Download all runs to a file. By default uses the REST API; pass ",{"type":53,"tag":74,"props":1138,"children":1140},{"className":1139},[],[1141],{"type":59,"value":1142},"--all",{"type":59,"value":1144}," to use Arrow Flight for bulk transfer.",{"type":53,"tag":386,"props":1146,"children":1148},{"className":388,"code":1147,"language":390,"meta":391,"style":391},"# EXPERIMENT_NAME, DATASET_NAME: name or ID (name preferred)\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE\n# -> experiment_abc123_20260305_141500\u002Fruns.json\n\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --all\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --output-dir .\u002Fresults\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq '.[0]'\n",[1149],{"type":53,"tag":74,"props":1150,"children":1151},{"__ignoreMap":391},[1152,1160,1198,1206,1215,1255,1301,1342],{"type":53,"tag":397,"props":1153,"children":1154},{"class":399,"line":400},[1155],{"type":53,"tag":397,"props":1156,"children":1157},{"style":468},[1158],{"type":59,"value":1159},"# EXPERIMENT_NAME, DATASET_NAME: name or ID (name preferred)\n",{"type":53,"tag":397,"props":1161,"children":1162},{"class":399,"line":420},[1163,1167,1171,1176,1181,1185,1189,1193],{"type":53,"tag":397,"props":1164,"children":1165},{"style":404},[1166],{"type":59,"value":224},{"type":53,"tag":397,"props":1168,"children":1169},{"style":409},[1170],{"type":59,"value":412},{"type":53,"tag":397,"props":1172,"children":1173},{"style":409},[1174],{"type":59,"value":1175}," export",{"type":53,"tag":397,"props":1177,"children":1178},{"style":409},[1179],{"type":59,"value":1180}," EXPERIMENT_NAME",{"type":53,"tag":397,"props":1182,"children":1183},{"style":409},[1184],{"type":59,"value":439},{"type":53,"tag":397,"props":1186,"children":1187},{"style":409},[1188],{"type":59,"value":444},{"type":53,"tag":397,"props":1190,"children":1191},{"style":409},[1192],{"type":59,"value":449},{"type":53,"tag":397,"props":1194,"children":1195},{"style":409},[1196],{"type":59,"value":1197}," SPACE\n",{"type":53,"tag":397,"props":1199,"children":1200},{"class":399,"line":474},[1201],{"type":53,"tag":397,"props":1202,"children":1203},{"style":468},[1204],{"type":59,"value":1205},"# -> experiment_abc123_20260305_141500\u002Fruns.json\n",{"type":53,"tag":397,"props":1207,"children":1208},{"class":399,"line":500},[1209],{"type":53,"tag":397,"props":1210,"children":1212},{"emptyLinePlaceholder":1211},true,[1213],{"type":59,"value":1214},"\n",{"type":53,"tag":397,"props":1216,"children":1217},{"class":399,"line":27},[1218,1222,1226,1230,1234,1238,1242,1246,1250],{"type":53,"tag":397,"props":1219,"children":1220},{"style":404},[1221],{"type":59,"value":224},{"type":53,"tag":397,"props":1223,"children":1224},{"style":409},[1225],{"type":59,"value":412},{"type":53,"tag":397,"props":1227,"children":1228},{"style":409},[1229],{"type":59,"value":1175},{"type":53,"tag":397,"props":1231,"children":1232},{"style":409},[1233],{"type":59,"value":1180},{"type":53,"tag":397,"props":1235,"children":1236},{"style":409},[1237],{"type":59,"value":439},{"type":53,"tag":397,"props":1239,"children":1240},{"style":409},[1241],{"type":59,"value":444},{"type":53,"tag":397,"props":1243,"children":1244},{"style":409},[1245],{"type":59,"value":449},{"type":53,"tag":397,"props":1247,"children":1248},{"style":409},[1249],{"type":59,"value":454},{"type":53,"tag":397,"props":1251,"children":1252},{"style":409},[1253],{"type":59,"value":1254}," --all\n",{"type":53,"tag":397,"props":1256,"children":1258},{"class":399,"line":1257},6,[1259,1263,1267,1271,1275,1279,1283,1287,1291,1296],{"type":53,"tag":397,"props":1260,"children":1261},{"style":404},[1262],{"type":59,"value":224},{"type":53,"tag":397,"props":1264,"children":1265},{"style":409},[1266],{"type":59,"value":412},{"type":53,"tag":397,"props":1268,"children":1269},{"style":409},[1270],{"type":59,"value":1175},{"type":53,"tag":397,"props":1272,"children":1273},{"style":409},[1274],{"type":59,"value":1180},{"type":53,"tag":397,"props":1276,"children":1277},{"style":409},[1278],{"type":59,"value":439},{"type":53,"tag":397,"props":1280,"children":1281},{"style":409},[1282],{"type":59,"value":444},{"type":53,"tag":397,"props":1284,"children":1285},{"style":409},[1286],{"type":59,"value":449},{"type":53,"tag":397,"props":1288,"children":1289},{"style":409},[1290],{"type":59,"value":454},{"type":53,"tag":397,"props":1292,"children":1293},{"style":409},[1294],{"type":59,"value":1295}," --output-dir",{"type":53,"tag":397,"props":1297,"children":1298},{"style":409},[1299],{"type":59,"value":1300}," .\u002Fresults\n",{"type":53,"tag":397,"props":1302,"children":1304},{"class":399,"line":1303},7,[1305,1309,1313,1317,1321,1325,1329,1333,1337],{"type":53,"tag":397,"props":1306,"children":1307},{"style":404},[1308],{"type":59,"value":224},{"type":53,"tag":397,"props":1310,"children":1311},{"style":409},[1312],{"type":59,"value":412},{"type":53,"tag":397,"props":1314,"children":1315},{"style":409},[1316],{"type":59,"value":1175},{"type":53,"tag":397,"props":1318,"children":1319},{"style":409},[1320],{"type":59,"value":1180},{"type":53,"tag":397,"props":1322,"children":1323},{"style":409},[1324],{"type":59,"value":439},{"type":53,"tag":397,"props":1326,"children":1327},{"style":409},[1328],{"type":59,"value":444},{"type":53,"tag":397,"props":1330,"children":1331},{"style":409},[1332],{"type":59,"value":449},{"type":53,"tag":397,"props":1334,"children":1335},{"style":409},[1336],{"type":59,"value":454},{"type":53,"tag":397,"props":1338,"children":1339},{"style":409},[1340],{"type":59,"value":1341}," --stdout\n",{"type":53,"tag":397,"props":1343,"children":1345},{"class":399,"line":1344},8,[1346,1350,1354,1358,1362,1366,1370,1374,1378,1383,1389,1394,1399,1404],{"type":53,"tag":397,"props":1347,"children":1348},{"style":404},[1349],{"type":59,"value":224},{"type":53,"tag":397,"props":1351,"children":1352},{"style":409},[1353],{"type":59,"value":412},{"type":53,"tag":397,"props":1355,"children":1356},{"style":409},[1357],{"type":59,"value":1175},{"type":53,"tag":397,"props":1359,"children":1360},{"style":409},[1361],{"type":59,"value":1180},{"type":53,"tag":397,"props":1363,"children":1364},{"style":409},[1365],{"type":59,"value":439},{"type":53,"tag":397,"props":1367,"children":1368},{"style":409},[1369],{"type":59,"value":444},{"type":53,"tag":397,"props":1371,"children":1372},{"style":409},[1373],{"type":59,"value":449},{"type":53,"tag":397,"props":1375,"children":1376},{"style":409},[1377],{"type":59,"value":454},{"type":53,"tag":397,"props":1379,"children":1380},{"style":409},[1381],{"type":59,"value":1382}," --stdout",{"type":53,"tag":397,"props":1384,"children":1386},{"style":1385},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[1387],{"type":59,"value":1388}," |",{"type":53,"tag":397,"props":1390,"children":1391},{"style":404},[1392],{"type":59,"value":1393}," jq",{"type":53,"tag":397,"props":1395,"children":1396},{"style":1385},[1397],{"type":59,"value":1398}," '",{"type":53,"tag":397,"props":1400,"children":1401},{"style":409},[1402],{"type":59,"value":1403},".[0]",{"type":53,"tag":397,"props":1405,"children":1406},{"style":1385},[1407],{"type":59,"value":1408},"'\n",{"type":53,"tag":525,"props":1410,"children":1412},{"id":1411},"flags-2",[1413],{"type":59,"value":530},{"type":53,"tag":532,"props":1415,"children":1416},{},[1417,1439],{"type":53,"tag":536,"props":1418,"children":1419},{},[1420],{"type":53,"tag":540,"props":1421,"children":1422},{},[1423,1427,1431,1435],{"type":53,"tag":544,"props":1424,"children":1425},{},[1426],{"type":59,"value":548},{"type":53,"tag":544,"props":1428,"children":1429},{},[1430],{"type":59,"value":553},{"type":53,"tag":544,"props":1432,"children":1433},{},[1434],{"type":59,"value":558},{"type":53,"tag":544,"props":1436,"children":1437},{},[1438],{"type":59,"value":563},{"type":53,"tag":565,"props":1440,"children":1441},{},[1442,1465,1488,1511,1537,1566],{"type":53,"tag":540,"props":1443,"children":1444},{},[1445,1453,1457,1461],{"type":53,"tag":572,"props":1446,"children":1447},{},[1448],{"type":53,"tag":74,"props":1449,"children":1451},{"className":1450},[],[1452],{"type":59,"value":855},{"type":53,"tag":572,"props":1454,"children":1455},{},[1456],{"type":59,"value":585},{"type":53,"tag":572,"props":1458,"children":1459},{},[1460],{"type":59,"value":864},{"type":53,"tag":572,"props":1462,"children":1463},{},[1464],{"type":59,"value":869},{"type":53,"tag":540,"props":1466,"children":1467},{},[1468,1476,1480,1484],{"type":53,"tag":572,"props":1469,"children":1470},{},[1471],{"type":53,"tag":74,"props":1472,"children":1474},{"className":1473},[],[1475],{"type":59,"value":580},{"type":53,"tag":572,"props":1477,"children":1478},{},[1479],{"type":59,"value":585},{"type":53,"tag":572,"props":1481,"children":1482},{},[1483],{"type":59,"value":590},{"type":53,"tag":572,"props":1485,"children":1486},{},[1487],{"type":59,"value":893},{"type":53,"tag":540,"props":1489,"children":1490},{},[1491,1499,1503,1507],{"type":53,"tag":572,"props":1492,"children":1493},{},[1494],{"type":53,"tag":74,"props":1495,"children":1497},{"className":1496},[],[1498],{"type":59,"value":87},{"type":53,"tag":572,"props":1500,"children":1501},{},[1502],{"type":59,"value":585},{"type":53,"tag":572,"props":1504,"children":1505},{},[1506],{"type":59,"value":590},{"type":53,"tag":572,"props":1508,"children":1509},{},[1510],{"type":59,"value":917},{"type":53,"tag":540,"props":1512,"children":1513},{},[1514,1522,1527,1532],{"type":53,"tag":572,"props":1515,"children":1516},{},[1517],{"type":53,"tag":74,"props":1518,"children":1520},{"className":1519},[],[1521],{"type":59,"value":1142},{"type":53,"tag":572,"props":1523,"children":1524},{},[1525],{"type":59,"value":1526},"bool",{"type":53,"tag":572,"props":1528,"children":1529},{},[1530],{"type":59,"value":1531},"false",{"type":53,"tag":572,"props":1533,"children":1534},{},[1535],{"type":59,"value":1536},"Use Arrow Flight for bulk export (see below)",{"type":53,"tag":540,"props":1538,"children":1539},{},[1540,1549,1553,1561],{"type":53,"tag":572,"props":1541,"children":1542},{},[1543],{"type":53,"tag":74,"props":1544,"children":1546},{"className":1545},[],[1547],{"type":59,"value":1548},"--output-dir",{"type":53,"tag":572,"props":1550,"children":1551},{},[1552],{"type":59,"value":585},{"type":53,"tag":572,"props":1554,"children":1555},{},[1556],{"type":53,"tag":74,"props":1557,"children":1559},{"className":1558},[],[1560],{"type":59,"value":134},{"type":53,"tag":572,"props":1562,"children":1563},{},[1564],{"type":59,"value":1565},"Output directory",{"type":53,"tag":540,"props":1567,"children":1568},{},[1569,1578,1582,1586],{"type":53,"tag":572,"props":1570,"children":1571},{},[1572],{"type":53,"tag":74,"props":1573,"children":1575},{"className":1574},[],[1576],{"type":59,"value":1577},"--stdout",{"type":53,"tag":572,"props":1579,"children":1580},{},[1581],{"type":59,"value":1526},{"type":53,"tag":572,"props":1583,"children":1584},{},[1585],{"type":59,"value":1531},{"type":53,"tag":572,"props":1587,"children":1588},{},[1589],{"type":59,"value":1590},"Print JSON to stdout instead of file",{"type":53,"tag":525,"props":1592,"children":1594},{"id":1593},"rest-vs-flight-all",[1595,1597,1602],{"type":59,"value":1596},"REST vs Flight (",{"type":53,"tag":74,"props":1598,"children":1600},{"className":1599},[],[1601],{"type":59,"value":1142},{"type":59,"value":1603},")",{"type":53,"tag":143,"props":1605,"children":1606},{},[1607,1617],{"type":53,"tag":147,"props":1608,"children":1609},{},[1610,1615],{"type":53,"tag":70,"props":1611,"children":1612},{},[1613],{"type":59,"value":1614},"REST",{"type":59,"value":1616}," (default): Lower friction -- no Arrow\u002FFlight dependency, standard HTTPS ports, works through any corporate proxy or firewall. Limited to 500 runs per page.",{"type":53,"tag":147,"props":1618,"children":1619},{},[1620,1625,1627,1632,1634,1639,1641,1646],{"type":53,"tag":70,"props":1621,"children":1622},{},[1623],{"type":59,"value":1624},"Flight",{"type":59,"value":1626}," (",{"type":53,"tag":74,"props":1628,"children":1630},{"className":1629},[],[1631],{"type":59,"value":1142},{"type":59,"value":1633},"): Required for experiments with more than 500 runs. Uses gRPC+TLS on a separate host\u002Fport which some corporate networks may block. The active ",{"type":53,"tag":74,"props":1635,"children":1637},{"className":1636},[],[1638],{"type":59,"value":224},{"type":59,"value":1640}," profile supplies the regional endpoint; see ",{"type":53,"tag":254,"props":1642,"children":1643},{"href":280},[1644],{"type":59,"value":1645},"profile setup",{"type":59,"value":134},{"type":53,"tag":66,"props":1648,"children":1649},{},[1650,1655,1657,1662],{"type":53,"tag":70,"props":1651,"children":1652},{},[1653],{"type":59,"value":1654},"Agent auto-escalation rule:",{"type":59,"value":1656}," If a REST export returns exactly 500 runs, the result is likely truncated. Re-run with ",{"type":53,"tag":74,"props":1658,"children":1660},{"className":1659},[],[1661],{"type":59,"value":1142},{"type":59,"value":1663}," to get the full dataset.",{"type":53,"tag":66,"props":1665,"children":1666},{},[1667],{"type":59,"value":1668},"Output is a JSON array of run objects:",{"type":53,"tag":386,"props":1670,"children":1674},{"className":1671,"code":1672,"language":1673,"meta":391,"style":391},"language-json shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","[\n  {\n    \"id\": \"run_001\",\n    \"example_id\": \"ex_001\",\n    \"output\": \"The answer is 4.\",\n    \"evaluations\": {\n      \"correctness\": { \"label\": \"correct\", \"score\": 1.0 },\n      \"relevance\": { \"score\": 0.95, \"explanation\": \"Directly answers the question\" }\n    },\n    \"metadata\": { \"model\": \"gpt-4o\", \"latency_ms\": 1234 }\n  }\n]\n","json",[1675],{"type":53,"tag":74,"props":1676,"children":1677},{"__ignoreMap":391},[1678,1686,1694,1736,1773,1810,1835,1922,2005,2014,2099,2108],{"type":53,"tag":397,"props":1679,"children":1680},{"class":399,"line":400},[1681],{"type":53,"tag":397,"props":1682,"children":1683},{"style":1385},[1684],{"type":59,"value":1685},"[\n",{"type":53,"tag":397,"props":1687,"children":1688},{"class":399,"line":420},[1689],{"type":53,"tag":397,"props":1690,"children":1691},{"style":1385},[1692],{"type":59,"value":1693},"  {\n",{"type":53,"tag":397,"props":1695,"children":1696},{"class":399,"line":474},[1697,1702,1707,1712,1717,1722,1727,1731],{"type":53,"tag":397,"props":1698,"children":1699},{"style":1385},[1700],{"type":59,"value":1701},"    \"",{"type":53,"tag":397,"props":1703,"children":1705},{"style":1704},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[1706],{"type":59,"value":984},{"type":53,"tag":397,"props":1708,"children":1709},{"style":1385},[1710],{"type":59,"value":1711},"\"",{"type":53,"tag":397,"props":1713,"children":1714},{"style":1385},[1715],{"type":59,"value":1716},":",{"type":53,"tag":397,"props":1718,"children":1719},{"style":1385},[1720],{"type":59,"value":1721}," \"",{"type":53,"tag":397,"props":1723,"children":1724},{"style":409},[1725],{"type":59,"value":1726},"run_001",{"type":53,"tag":397,"props":1728,"children":1729},{"style":1385},[1730],{"type":59,"value":1711},{"type":53,"tag":397,"props":1732,"children":1733},{"style":1385},[1734],{"type":59,"value":1735},",\n",{"type":53,"tag":397,"props":1737,"children":1738},{"class":399,"line":500},[1739,1743,1748,1752,1756,1760,1765,1769],{"type":53,"tag":397,"props":1740,"children":1741},{"style":1385},[1742],{"type":59,"value":1701},{"type":53,"tag":397,"props":1744,"children":1745},{"style":1704},[1746],{"type":59,"value":1747},"example_id",{"type":53,"tag":397,"props":1749,"children":1750},{"style":1385},[1751],{"type":59,"value":1711},{"type":53,"tag":397,"props":1753,"children":1754},{"style":1385},[1755],{"type":59,"value":1716},{"type":53,"tag":397,"props":1757,"children":1758},{"style":1385},[1759],{"type":59,"value":1721},{"type":53,"tag":397,"props":1761,"children":1762},{"style":409},[1763],{"type":59,"value":1764},"ex_001",{"type":53,"tag":397,"props":1766,"children":1767},{"style":1385},[1768],{"type":59,"value":1711},{"type":53,"tag":397,"props":1770,"children":1771},{"style":1385},[1772],{"type":59,"value":1735},{"type":53,"tag":397,"props":1774,"children":1775},{"class":399,"line":27},[1776,1780,1785,1789,1793,1797,1802,1806],{"type":53,"tag":397,"props":1777,"children":1778},{"style":1385},[1779],{"type":59,"value":1701},{"type":53,"tag":397,"props":1781,"children":1782},{"style":1704},[1783],{"type":59,"value":1784},"output",{"type":53,"tag":397,"props":1786,"children":1787},{"style":1385},[1788],{"type":59,"value":1711},{"type":53,"tag":397,"props":1790,"children":1791},{"style":1385},[1792],{"type":59,"value":1716},{"type":53,"tag":397,"props":1794,"children":1795},{"style":1385},[1796],{"type":59,"value":1721},{"type":53,"tag":397,"props":1798,"children":1799},{"style":409},[1800],{"type":59,"value":1801},"The answer is 4.",{"type":53,"tag":397,"props":1803,"children":1804},{"style":1385},[1805],{"type":59,"value":1711},{"type":53,"tag":397,"props":1807,"children":1808},{"style":1385},[1809],{"type":59,"value":1735},{"type":53,"tag":397,"props":1811,"children":1812},{"class":399,"line":1257},[1813,1817,1822,1826,1830],{"type":53,"tag":397,"props":1814,"children":1815},{"style":1385},[1816],{"type":59,"value":1701},{"type":53,"tag":397,"props":1818,"children":1819},{"style":1704},[1820],{"type":59,"value":1821},"evaluations",{"type":53,"tag":397,"props":1823,"children":1824},{"style":1385},[1825],{"type":59,"value":1711},{"type":53,"tag":397,"props":1827,"children":1828},{"style":1385},[1829],{"type":59,"value":1716},{"type":53,"tag":397,"props":1831,"children":1832},{"style":1385},[1833],{"type":59,"value":1834}," {\n",{"type":53,"tag":397,"props":1836,"children":1837},{"class":399,"line":1303},[1838,1843,1847,1851,1855,1860,1864,1869,1873,1877,1881,1886,1890,1895,1899,1904,1908,1912,1917],{"type":53,"tag":397,"props":1839,"children":1840},{"style":1385},[1841],{"type":59,"value":1842},"      \"",{"type":53,"tag":397,"props":1844,"children":1845},{"style":404},[1846],{"type":59,"value":192},{"type":53,"tag":397,"props":1848,"children":1849},{"style":1385},[1850],{"type":59,"value":1711},{"type":53,"tag":397,"props":1852,"children":1853},{"style":1385},[1854],{"type":59,"value":1716},{"type":53,"tag":397,"props":1856,"children":1857},{"style":1385},[1858],{"type":59,"value":1859}," {",{"type":53,"tag":397,"props":1861,"children":1862},{"style":1385},[1863],{"type":59,"value":1721},{"type":53,"tag":397,"props":1865,"children":1866},{"style":462},[1867],{"type":59,"value":1868},"label",{"type":53,"tag":397,"props":1870,"children":1871},{"style":1385},[1872],{"type":59,"value":1711},{"type":53,"tag":397,"props":1874,"children":1875},{"style":1385},[1876],{"type":59,"value":1716},{"type":53,"tag":397,"props":1878,"children":1879},{"style":1385},[1880],{"type":59,"value":1721},{"type":53,"tag":397,"props":1882,"children":1883},{"style":409},[1884],{"type":59,"value":1885},"correct",{"type":53,"tag":397,"props":1887,"children":1888},{"style":1385},[1889],{"type":59,"value":1711},{"type":53,"tag":397,"props":1891,"children":1892},{"style":1385},[1893],{"type":59,"value":1894},",",{"type":53,"tag":397,"props":1896,"children":1897},{"style":1385},[1898],{"type":59,"value":1721},{"type":53,"tag":397,"props":1900,"children":1901},{"style":462},[1902],{"type":59,"value":1903},"score",{"type":53,"tag":397,"props":1905,"children":1906},{"style":1385},[1907],{"type":59,"value":1711},{"type":53,"tag":397,"props":1909,"children":1910},{"style":1385},[1911],{"type":59,"value":1716},{"type":53,"tag":397,"props":1913,"children":1914},{"style":462},[1915],{"type":59,"value":1916}," 1.0",{"type":53,"tag":397,"props":1918,"children":1919},{"style":1385},[1920],{"type":59,"value":1921}," },\n",{"type":53,"tag":397,"props":1923,"children":1924},{"class":399,"line":1344},[1925,1929,1933,1937,1941,1945,1949,1953,1957,1961,1966,1970,1974,1979,1983,1987,1991,1996,2000],{"type":53,"tag":397,"props":1926,"children":1927},{"style":1385},[1928],{"type":59,"value":1842},{"type":53,"tag":397,"props":1930,"children":1931},{"style":404},[1932],{"type":59,"value":200},{"type":53,"tag":397,"props":1934,"children":1935},{"style":1385},[1936],{"type":59,"value":1711},{"type":53,"tag":397,"props":1938,"children":1939},{"style":1385},[1940],{"type":59,"value":1716},{"type":53,"tag":397,"props":1942,"children":1943},{"style":1385},[1944],{"type":59,"value":1859},{"type":53,"tag":397,"props":1946,"children":1947},{"style":1385},[1948],{"type":59,"value":1721},{"type":53,"tag":397,"props":1950,"children":1951},{"style":462},[1952],{"type":59,"value":1903},{"type":53,"tag":397,"props":1954,"children":1955},{"style":1385},[1956],{"type":59,"value":1711},{"type":53,"tag":397,"props":1958,"children":1959},{"style":1385},[1960],{"type":59,"value":1716},{"type":53,"tag":397,"props":1962,"children":1963},{"style":462},[1964],{"type":59,"value":1965}," 0.95",{"type":53,"tag":397,"props":1967,"children":1968},{"style":1385},[1969],{"type":59,"value":1894},{"type":53,"tag":397,"props":1971,"children":1972},{"style":1385},[1973],{"type":59,"value":1721},{"type":53,"tag":397,"props":1975,"children":1976},{"style":462},[1977],{"type":59,"value":1978},"explanation",{"type":53,"tag":397,"props":1980,"children":1981},{"style":1385},[1982],{"type":59,"value":1711},{"type":53,"tag":397,"props":1984,"children":1985},{"style":1385},[1986],{"type":59,"value":1716},{"type":53,"tag":397,"props":1988,"children":1989},{"style":1385},[1990],{"type":59,"value":1721},{"type":53,"tag":397,"props":1992,"children":1993},{"style":409},[1994],{"type":59,"value":1995},"Directly answers the question",{"type":53,"tag":397,"props":1997,"children":1998},{"style":1385},[1999],{"type":59,"value":1711},{"type":53,"tag":397,"props":2001,"children":2002},{"style":1385},[2003],{"type":59,"value":2004}," }\n",{"type":53,"tag":397,"props":2006,"children":2008},{"class":399,"line":2007},9,[2009],{"type":53,"tag":397,"props":2010,"children":2011},{"style":1385},[2012],{"type":59,"value":2013},"    },\n",{"type":53,"tag":397,"props":2015,"children":2017},{"class":399,"line":2016},10,[2018,2022,2027,2031,2035,2039,2043,2048,2052,2056,2060,2065,2069,2073,2077,2082,2086,2090,2095],{"type":53,"tag":397,"props":2019,"children":2020},{"style":1385},[2021],{"type":59,"value":1701},{"type":53,"tag":397,"props":2023,"children":2024},{"style":1704},[2025],{"type":59,"value":2026},"metadata",{"type":53,"tag":397,"props":2028,"children":2029},{"style":1385},[2030],{"type":59,"value":1711},{"type":53,"tag":397,"props":2032,"children":2033},{"style":1385},[2034],{"type":59,"value":1716},{"type":53,"tag":397,"props":2036,"children":2037},{"style":1385},[2038],{"type":59,"value":1859},{"type":53,"tag":397,"props":2040,"children":2041},{"style":1385},[2042],{"type":59,"value":1721},{"type":53,"tag":397,"props":2044,"children":2045},{"style":404},[2046],{"type":59,"value":2047},"model",{"type":53,"tag":397,"props":2049,"children":2050},{"style":1385},[2051],{"type":59,"value":1711},{"type":53,"tag":397,"props":2053,"children":2054},{"style":1385},[2055],{"type":59,"value":1716},{"type":53,"tag":397,"props":2057,"children":2058},{"style":1385},[2059],{"type":59,"value":1721},{"type":53,"tag":397,"props":2061,"children":2062},{"style":409},[2063],{"type":59,"value":2064},"gpt-4o",{"type":53,"tag":397,"props":2066,"children":2067},{"style":1385},[2068],{"type":59,"value":1711},{"type":53,"tag":397,"props":2070,"children":2071},{"style":1385},[2072],{"type":59,"value":1894},{"type":53,"tag":397,"props":2074,"children":2075},{"style":1385},[2076],{"type":59,"value":1721},{"type":53,"tag":397,"props":2078,"children":2079},{"style":404},[2080],{"type":59,"value":2081},"latency_ms",{"type":53,"tag":397,"props":2083,"children":2084},{"style":1385},[2085],{"type":59,"value":1711},{"type":53,"tag":397,"props":2087,"children":2088},{"style":1385},[2089],{"type":59,"value":1716},{"type":53,"tag":397,"props":2091,"children":2092},{"style":462},[2093],{"type":59,"value":2094}," 1234",{"type":53,"tag":397,"props":2096,"children":2097},{"style":1385},[2098],{"type":59,"value":2004},{"type":53,"tag":397,"props":2100,"children":2102},{"class":399,"line":2101},11,[2103],{"type":53,"tag":397,"props":2104,"children":2105},{"style":1385},[2106],{"type":59,"value":2107},"  }\n",{"type":53,"tag":397,"props":2109,"children":2111},{"class":399,"line":2110},12,[2112],{"type":53,"tag":397,"props":2113,"children":2114},{"style":1385},[2115],{"type":59,"value":2116},"]\n",{"type":53,"tag":136,"props":2118,"children":2120},{"id":2119},"create-experiment-ax-experiments-create",[2121,2123],{"type":59,"value":2122},"Create Experiment: ",{"type":53,"tag":74,"props":2124,"children":2126},{"className":2125},[],[2127],{"type":59,"value":2128},"ax experiments create",{"type":53,"tag":66,"props":2130,"children":2131},{},[2132],{"type":59,"value":2133},"Create a new experiment with runs from a data file.",{"type":53,"tag":386,"props":2135,"children":2137},{"className":388,"code":2136,"language":390,"meta":391,"style":391},"ax experiments create --name \"gpt-4o-baseline\" --dataset DATASET_NAME --space SPACE --file runs.json\nax experiments create --name \"claude-test\" --dataset DATASET_NAME --space SPACE --file runs.csv\n",[2138],{"type":53,"tag":74,"props":2139,"children":2140},{"__ignoreMap":391},[2141,2201],{"type":53,"tag":397,"props":2142,"children":2143},{"class":399,"line":400},[2144,2148,2152,2157,2162,2166,2171,2175,2179,2183,2187,2191,2196],{"type":53,"tag":397,"props":2145,"children":2146},{"style":404},[2147],{"type":59,"value":224},{"type":53,"tag":397,"props":2149,"children":2150},{"style":409},[2151],{"type":59,"value":412},{"type":53,"tag":397,"props":2153,"children":2154},{"style":409},[2155],{"type":59,"value":2156}," create",{"type":53,"tag":397,"props":2158,"children":2159},{"style":409},[2160],{"type":59,"value":2161}," --name",{"type":53,"tag":397,"props":2163,"children":2164},{"style":1385},[2165],{"type":59,"value":1721},{"type":53,"tag":397,"props":2167,"children":2168},{"style":409},[2169],{"type":59,"value":2170},"gpt-4o-baseline",{"type":53,"tag":397,"props":2172,"children":2173},{"style":1385},[2174],{"type":59,"value":1711},{"type":53,"tag":397,"props":2176,"children":2177},{"style":409},[2178],{"type":59,"value":439},{"type":53,"tag":397,"props":2180,"children":2181},{"style":409},[2182],{"type":59,"value":444},{"type":53,"tag":397,"props":2184,"children":2185},{"style":409},[2186],{"type":59,"value":449},{"type":53,"tag":397,"props":2188,"children":2189},{"style":409},[2190],{"type":59,"value":454},{"type":53,"tag":397,"props":2192,"children":2193},{"style":409},[2194],{"type":59,"value":2195}," --file",{"type":53,"tag":397,"props":2197,"children":2198},{"style":409},[2199],{"type":59,"value":2200}," runs.json\n",{"type":53,"tag":397,"props":2202,"children":2203},{"class":399,"line":420},[2204,2208,2212,2216,2220,2224,2229,2233,2237,2241,2245,2249,2253],{"type":53,"tag":397,"props":2205,"children":2206},{"style":404},[2207],{"type":59,"value":224},{"type":53,"tag":397,"props":2209,"children":2210},{"style":409},[2211],{"type":59,"value":412},{"type":53,"tag":397,"props":2213,"children":2214},{"style":409},[2215],{"type":59,"value":2156},{"type":53,"tag":397,"props":2217,"children":2218},{"style":409},[2219],{"type":59,"value":2161},{"type":53,"tag":397,"props":2221,"children":2222},{"style":1385},[2223],{"type":59,"value":1721},{"type":53,"tag":397,"props":2225,"children":2226},{"style":409},[2227],{"type":59,"value":2228},"claude-test",{"type":53,"tag":397,"props":2230,"children":2231},{"style":1385},[2232],{"type":59,"value":1711},{"type":53,"tag":397,"props":2234,"children":2235},{"style":409},[2236],{"type":59,"value":439},{"type":53,"tag":397,"props":2238,"children":2239},{"style":409},[2240],{"type":59,"value":444},{"type":53,"tag":397,"props":2242,"children":2243},{"style":409},[2244],{"type":59,"value":449},{"type":53,"tag":397,"props":2246,"children":2247},{"style":409},[2248],{"type":59,"value":454},{"type":53,"tag":397,"props":2250,"children":2251},{"style":409},[2252],{"type":59,"value":2195},{"type":53,"tag":397,"props":2254,"children":2255},{"style":409},[2256],{"type":59,"value":2257}," runs.csv\n",{"type":53,"tag":525,"props":2259,"children":2261},{"id":2260},"flags-3",[2262],{"type":59,"value":530},{"type":53,"tag":532,"props":2264,"children":2265},{},[2266,2289],{"type":53,"tag":536,"props":2267,"children":2268},{},[2269],{"type":53,"tag":540,"props":2270,"children":2271},{},[2272,2276,2280,2285],{"type":53,"tag":544,"props":2273,"children":2274},{},[2275],{"type":59,"value":548},{"type":53,"tag":544,"props":2277,"children":2278},{},[2279],{"type":59,"value":553},{"type":53,"tag":544,"props":2281,"children":2282},{},[2283],{"type":59,"value":2284},"Required",{"type":53,"tag":544,"props":2286,"children":2287},{},[2288],{"type":59,"value":563},{"type":53,"tag":565,"props":2290,"children":2291},{},[2292,2316,2340,2365,2391],{"type":53,"tag":540,"props":2293,"children":2294},{},[2295,2303,2307,2312],{"type":53,"tag":572,"props":2296,"children":2297},{},[2298],{"type":53,"tag":74,"props":2299,"children":2301},{"className":2300},[],[2302],{"type":59,"value":607},{"type":53,"tag":572,"props":2304,"children":2305},{},[2306],{"type":59,"value":585},{"type":53,"tag":572,"props":2308,"children":2309},{},[2310],{"type":59,"value":2311},"yes",{"type":53,"tag":572,"props":2313,"children":2314},{},[2315],{"type":59,"value":1013},{"type":53,"tag":540,"props":2317,"children":2318},{},[2319,2327,2331,2335],{"type":53,"tag":572,"props":2320,"children":2321},{},[2322],{"type":53,"tag":74,"props":2323,"children":2325},{"className":2324},[],[2326],{"type":59,"value":580},{"type":53,"tag":572,"props":2328,"children":2329},{},[2330],{"type":59,"value":585},{"type":53,"tag":572,"props":2332,"children":2333},{},[2334],{"type":59,"value":2311},{"type":53,"tag":572,"props":2336,"children":2337},{},[2338],{"type":59,"value":2339},"Dataset to run the experiment against",{"type":53,"tag":540,"props":2341,"children":2342},{},[2343,2352,2356,2361],{"type":53,"tag":572,"props":2344,"children":2345},{},[2346],{"type":53,"tag":74,"props":2347,"children":2349},{"className":2348},[],[2350],{"type":59,"value":2351},"--space, -s",{"type":53,"tag":572,"props":2353,"children":2354},{},[2355],{"type":59,"value":585},{"type":53,"tag":572,"props":2357,"children":2358},{},[2359],{"type":59,"value":2360},"no",{"type":53,"tag":572,"props":2362,"children":2363},{},[2364],{"type":59,"value":917},{"type":53,"tag":540,"props":2366,"children":2367},{},[2368,2377,2382,2386],{"type":53,"tag":572,"props":2369,"children":2370},{},[2371],{"type":53,"tag":74,"props":2372,"children":2374},{"className":2373},[],[2375],{"type":59,"value":2376},"--file, -f",{"type":53,"tag":572,"props":2378,"children":2379},{},[2380],{"type":59,"value":2381},"path",{"type":53,"tag":572,"props":2383,"children":2384},{},[2385],{"type":59,"value":2311},{"type":53,"tag":572,"props":2387,"children":2388},{},[2389],{"type":59,"value":2390},"Data file with runs: CSV, JSON, JSONL, or Parquet",{"type":53,"tag":540,"props":2392,"children":2393},{},[2394,2402,2406,2410],{"type":53,"tag":572,"props":2395,"children":2396},{},[2397],{"type":53,"tag":74,"props":2398,"children":2400},{"className":2399},[],[2401],{"type":59,"value":684},{"type":53,"tag":572,"props":2403,"children":2404},{},[2405],{"type":59,"value":585},{"type":53,"tag":572,"props":2407,"children":2408},{},[2409],{"type":59,"value":2360},{"type":53,"tag":572,"props":2411,"children":2412},{},[2413],{"type":59,"value":941},{"type":53,"tag":525,"props":2415,"children":2417},{"id":2416},"passing-data-via-stdin",[2418],{"type":59,"value":2419},"Passing data via stdin",{"type":53,"tag":66,"props":2421,"children":2422},{},[2423,2425,2431],{"type":59,"value":2424},"Use ",{"type":53,"tag":74,"props":2426,"children":2428},{"className":2427},[],[2429],{"type":59,"value":2430},"--file -",{"type":59,"value":2432}," to pipe data directly — no temp file needed:",{"type":53,"tag":386,"props":2434,"children":2436},{"className":388,"code":2435,"language":390,"meta":391,"style":391},"echo '[{\"example_id\": \"ex_001\", \"output\": \"Paris\"}]' | ax experiments create --name \"my-experiment\" --dataset DATASET_NAME --space SPACE --file -\n\n# Or with a heredoc\nax experiments create --name \"my-experiment\" --dataset DATASET_NAME --space SPACE --file - \u003C\u003C 'EOF'\n[{\"example_id\": \"ex_001\", \"output\": \"Paris\"}]\nEOF\n",[2437],{"type":53,"tag":74,"props":2438,"children":2439},{"__ignoreMap":391},[2440,2522,2529,2537,2603,2611],{"type":53,"tag":397,"props":2441,"children":2442},{"class":399,"line":400},[2443,2449,2453,2458,2463,2467,2472,2476,2480,2484,2488,2493,2497,2501,2505,2509,2513,2517],{"type":53,"tag":397,"props":2444,"children":2446},{"style":2445},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[2447],{"type":59,"value":2448},"echo",{"type":53,"tag":397,"props":2450,"children":2451},{"style":1385},[2452],{"type":59,"value":1398},{"type":53,"tag":397,"props":2454,"children":2455},{"style":409},[2456],{"type":59,"value":2457},"[{\"example_id\": \"ex_001\", \"output\": \"Paris\"}]",{"type":53,"tag":397,"props":2459,"children":2460},{"style":1385},[2461],{"type":59,"value":2462},"'",{"type":53,"tag":397,"props":2464,"children":2465},{"style":1385},[2466],{"type":59,"value":1388},{"type":53,"tag":397,"props":2468,"children":2469},{"style":404},[2470],{"type":59,"value":2471}," ax",{"type":53,"tag":397,"props":2473,"children":2474},{"style":409},[2475],{"type":59,"value":412},{"type":53,"tag":397,"props":2477,"children":2478},{"style":409},[2479],{"type":59,"value":2156},{"type":53,"tag":397,"props":2481,"children":2482},{"style":409},[2483],{"type":59,"value":2161},{"type":53,"tag":397,"props":2485,"children":2486},{"style":1385},[2487],{"type":59,"value":1721},{"type":53,"tag":397,"props":2489,"children":2490},{"style":409},[2491],{"type":59,"value":2492},"my-experiment",{"type":53,"tag":397,"props":2494,"children":2495},{"style":1385},[2496],{"type":59,"value":1711},{"type":53,"tag":397,"props":2498,"children":2499},{"style":409},[2500],{"type":59,"value":439},{"type":53,"tag":397,"props":2502,"children":2503},{"style":409},[2504],{"type":59,"value":444},{"type":53,"tag":397,"props":2506,"children":2507},{"style":409},[2508],{"type":59,"value":449},{"type":53,"tag":397,"props":2510,"children":2511},{"style":409},[2512],{"type":59,"value":454},{"type":53,"tag":397,"props":2514,"children":2515},{"style":409},[2516],{"type":59,"value":2195},{"type":53,"tag":397,"props":2518,"children":2519},{"style":409},[2520],{"type":59,"value":2521}," -\n",{"type":53,"tag":397,"props":2523,"children":2524},{"class":399,"line":420},[2525],{"type":53,"tag":397,"props":2526,"children":2527},{"emptyLinePlaceholder":1211},[2528],{"type":59,"value":1214},{"type":53,"tag":397,"props":2530,"children":2531},{"class":399,"line":474},[2532],{"type":53,"tag":397,"props":2533,"children":2534},{"style":468},[2535],{"type":59,"value":2536},"# Or with a heredoc\n",{"type":53,"tag":397,"props":2538,"children":2539},{"class":399,"line":500},[2540,2544,2548,2552,2556,2560,2564,2568,2572,2576,2580,2584,2588,2593,2598],{"type":53,"tag":397,"props":2541,"children":2542},{"style":404},[2543],{"type":59,"value":224},{"type":53,"tag":397,"props":2545,"children":2546},{"style":409},[2547],{"type":59,"value":412},{"type":53,"tag":397,"props":2549,"children":2550},{"style":409},[2551],{"type":59,"value":2156},{"type":53,"tag":397,"props":2553,"children":2554},{"style":409},[2555],{"type":59,"value":2161},{"type":53,"tag":397,"props":2557,"children":2558},{"style":1385},[2559],{"type":59,"value":1721},{"type":53,"tag":397,"props":2561,"children":2562},{"style":409},[2563],{"type":59,"value":2492},{"type":53,"tag":397,"props":2565,"children":2566},{"style":1385},[2567],{"type":59,"value":1711},{"type":53,"tag":397,"props":2569,"children":2570},{"style":409},[2571],{"type":59,"value":439},{"type":53,"tag":397,"props":2573,"children":2574},{"style":409},[2575],{"type":59,"value":444},{"type":53,"tag":397,"props":2577,"children":2578},{"style":409},[2579],{"type":59,"value":449},{"type":53,"tag":397,"props":2581,"children":2582},{"style":409},[2583],{"type":59,"value":454},{"type":53,"tag":397,"props":2585,"children":2586},{"style":409},[2587],{"type":59,"value":2195},{"type":53,"tag":397,"props":2589,"children":2590},{"style":409},[2591],{"type":59,"value":2592}," -",{"type":53,"tag":397,"props":2594,"children":2595},{"style":1385},[2596],{"type":59,"value":2597}," \u003C\u003C",{"type":53,"tag":397,"props":2599,"children":2600},{"style":1385},[2601],{"type":59,"value":2602}," 'EOF'\n",{"type":53,"tag":397,"props":2604,"children":2605},{"class":399,"line":27},[2606],{"type":53,"tag":397,"props":2607,"children":2608},{"style":409},[2609],{"type":59,"value":2610},"[{\"example_id\": \"ex_001\", \"output\": \"Paris\"}]\n",{"type":53,"tag":397,"props":2612,"children":2613},{"class":399,"line":1257},[2614],{"type":53,"tag":397,"props":2615,"children":2616},{"style":1385},[2617],{"type":59,"value":2618},"EOF\n",{"type":53,"tag":525,"props":2620,"children":2622},{"id":2621},"required-columns-in-the-runs-file",[2623],{"type":59,"value":2624},"Required columns in the runs file",{"type":53,"tag":532,"props":2626,"children":2627},{},[2628,2651],{"type":53,"tag":536,"props":2629,"children":2630},{},[2631],{"type":53,"tag":540,"props":2632,"children":2633},{},[2634,2639,2643,2647],{"type":53,"tag":544,"props":2635,"children":2636},{},[2637],{"type":59,"value":2638},"Column",{"type":53,"tag":544,"props":2640,"children":2641},{},[2642],{"type":59,"value":553},{"type":53,"tag":544,"props":2644,"children":2645},{},[2646],{"type":59,"value":2284},{"type":53,"tag":544,"props":2648,"children":2649},{},[2650],{"type":59,"value":563},{"type":53,"tag":565,"props":2652,"children":2653},{},[2654,2696],{"type":53,"tag":540,"props":2655,"children":2656},{},[2657,2665,2669,2673],{"type":53,"tag":572,"props":2658,"children":2659},{},[2660],{"type":53,"tag":74,"props":2661,"children":2663},{"className":2662},[],[2664],{"type":59,"value":1747},{"type":53,"tag":572,"props":2666,"children":2667},{},[2668],{"type":59,"value":585},{"type":53,"tag":572,"props":2670,"children":2671},{},[2672],{"type":59,"value":2311},{"type":53,"tag":572,"props":2674,"children":2675},{},[2676,2678,2688,2690],{"type":59,"value":2677},"The dataset example's ",{"type":53,"tag":70,"props":2679,"children":2680},{},[2681,2683],{"type":59,"value":2682},"top-level ",{"type":53,"tag":74,"props":2684,"children":2686},{"className":2685},[],[2687],{"type":59,"value":984},{"type":59,"value":2689}," from ",{"type":53,"tag":74,"props":2691,"children":2693},{"className":2692},[],[2694],{"type":59,"value":2695},"ax datasets export",{"type":53,"tag":540,"props":2697,"children":2698},{},[2699,2707,2711,2715],{"type":53,"tag":572,"props":2700,"children":2701},{},[2702],{"type":53,"tag":74,"props":2703,"children":2705},{"className":2704},[],[2706],{"type":59,"value":1784},{"type":53,"tag":572,"props":2708,"children":2709},{},[2710],{"type":59,"value":585},{"type":53,"tag":572,"props":2712,"children":2713},{},[2714],{"type":59,"value":2311},{"type":53,"tag":572,"props":2716,"children":2717},{},[2718],{"type":59,"value":2719},"The model\u002Fsystem output for this example",{"type":53,"tag":66,"props":2721,"children":2722},{},[2723,2725,2731],{"type":59,"value":2724},"Additional columns are passed through as ",{"type":53,"tag":74,"props":2726,"children":2728},{"className":2727},[],[2729],{"type":59,"value":2730},"additionalProperties",{"type":59,"value":2732}," on the run.",{"type":53,"tag":62,"props":2734,"children":2735},{},[2736],{"type":53,"tag":66,"props":2737,"children":2738},{},[2739,2749,2751,2756,2758,2764,2766,2771,2773,2779,2781,2786],{"type":53,"tag":70,"props":2740,"children":2741},{},[2742,2747],{"type":53,"tag":74,"props":2743,"children":2745},{"className":2744},[],[2746],{"type":59,"value":1747},{"type":59,"value":2748}," must be the Arize row id",{"type":59,"value":2750}," — the top-level ",{"type":53,"tag":74,"props":2752,"children":2754},{"className":2753},[],[2755],{"type":59,"value":984},{"type":59,"value":2757}," field on each exported dataset example (",{"type":53,"tag":74,"props":2759,"children":2761},{"className":2760},[],[2762],{"type":59,"value":2763},"ex[\"id\"]",{"type":59,"value":2765},"). Do ",{"type":53,"tag":70,"props":2767,"children":2768},{},[2769],{"type":59,"value":2770},"not",{"type":59,"value":2772}," use a value nested inside the example's input fields or ",{"type":53,"tag":74,"props":2774,"children":2776},{"className":2775},[],[2777],{"type":59,"value":2778},"additional_properties",{"type":59,"value":2780},"; a wrong value fails silently or attaches the run to the wrong example. Export the dataset and inspect the top-level ",{"type":53,"tag":74,"props":2782,"children":2784},{"className":2783},[],[2785],{"type":59,"value":984},{"type":59,"value":2787}," field before creating runs.",{"type":53,"tag":62,"props":2789,"children":2790},{},[2791],{"type":53,"tag":66,"props":2792,"children":2793},{},[2794,2799,2801,2807,2809,2814,2816,2821,2823,2828,2830,2834,2836,2840,2842,2848,2850,2855,2857,2862,2864,2869],{"type":53,"tag":70,"props":2795,"children":2796},{},[2797],{"type":59,"value":2798},"⚠️ Inline evaluations in the create file do NOT attach as scores.",{"type":59,"value":2800}," ",{"type":53,"tag":74,"props":2802,"children":2804},{"className":2803},[],[2805],{"type":59,"value":2806},"create",{"type":59,"value":2808}," only reads ",{"type":53,"tag":74,"props":2810,"children":2812},{"className":2811},[],[2813],{"type":59,"value":1747},{"type":59,"value":2815}," and ",{"type":53,"tag":74,"props":2817,"children":2819},{"className":2818},[],[2820],{"type":59,"value":1784},{"type":59,"value":2822},"; every other column — including an ",{"type":53,"tag":74,"props":2824,"children":2826},{"className":2825},[],[2827],{"type":59,"value":1821},{"type":59,"value":2829}," object — is stored as a passthrough additional field, ",{"type":53,"tag":70,"props":2831,"children":2832},{},[2833],{"type":59,"value":2770},{"type":59,"value":2835}," as an experiment evaluation, and will ",{"type":53,"tag":70,"props":2837,"children":2838},{},[2839],{"type":59,"value":2770},{"type":59,"value":2841}," appear as a score in the UI. This fails silently (no error). To attach scores\u002Flabels, create the experiment first, then run ",{"type":53,"tag":74,"props":2843,"children":2845},{"className":2844},[],[2846],{"type":59,"value":2847},"ax experiments annotate-runs",{"type":59,"value":2849},". The ",{"type":53,"tag":74,"props":2851,"children":2853},{"className":2852},[],[2854],{"type":59,"value":1821},{"type":59,"value":2856}," object in the schemas below is the ",{"type":53,"tag":70,"props":2858,"children":2859},{},[2860],{"type":59,"value":2861},"export (read)",{"type":59,"value":2863}," shape returned once annotations exist — it is not an input to ",{"type":53,"tag":74,"props":2865,"children":2867},{"className":2866},[],[2868],{"type":59,"value":2806},{"type":59,"value":134},{"type":53,"tag":136,"props":2871,"children":2873},{"id":2872},"delete-experiment-ax-experiments-delete",[2874,2876],{"type":59,"value":2875},"Delete Experiment: ",{"type":53,"tag":74,"props":2877,"children":2879},{"className":2878},[],[2880],{"type":59,"value":2881},"ax experiments delete",{"type":53,"tag":386,"props":2883,"children":2885},{"className":388,"code":2884,"language":390,"meta":391,"style":391},"ax experiments delete NAME_OR_ID\nax experiments delete NAME_OR_ID --dataset DATASET_NAME --space SPACE   # required when using experiment name instead of ID\nax experiments delete NAME_OR_ID --force   # skip confirmation prompt\n",[2886],{"type":53,"tag":74,"props":2887,"children":2888},{"__ignoreMap":391},[2889,2909,2948],{"type":53,"tag":397,"props":2890,"children":2891},{"class":399,"line":400},[2892,2896,2900,2905],{"type":53,"tag":397,"props":2893,"children":2894},{"style":404},[2895],{"type":59,"value":224},{"type":53,"tag":397,"props":2897,"children":2898},{"style":409},[2899],{"type":59,"value":412},{"type":53,"tag":397,"props":2901,"children":2902},{"style":409},[2903],{"type":59,"value":2904}," delete",{"type":53,"tag":397,"props":2906,"children":2907},{"style":409},[2908],{"type":59,"value":742},{"type":53,"tag":397,"props":2910,"children":2911},{"class":399,"line":420},[2912,2916,2920,2924,2928,2932,2936,2940,2944],{"type":53,"tag":397,"props":2913,"children":2914},{"style":404},[2915],{"type":59,"value":224},{"type":53,"tag":397,"props":2917,"children":2918},{"style":409},[2919],{"type":59,"value":412},{"type":53,"tag":397,"props":2921,"children":2922},{"style":409},[2923],{"type":59,"value":2904},{"type":53,"tag":397,"props":2925,"children":2926},{"style":409},[2927],{"type":59,"value":762},{"type":53,"tag":397,"props":2929,"children":2930},{"style":409},[2931],{"type":59,"value":439},{"type":53,"tag":397,"props":2933,"children":2934},{"style":409},[2935],{"type":59,"value":444},{"type":53,"tag":397,"props":2937,"children":2938},{"style":409},[2939],{"type":59,"value":449},{"type":53,"tag":397,"props":2941,"children":2942},{"style":409},[2943],{"type":59,"value":454},{"type":53,"tag":397,"props":2945,"children":2946},{"style":468},[2947],{"type":59,"value":810},{"type":53,"tag":397,"props":2949,"children":2950},{"class":399,"line":474},[2951,2955,2959,2963,2967,2972],{"type":53,"tag":397,"props":2952,"children":2953},{"style":404},[2954],{"type":59,"value":224},{"type":53,"tag":397,"props":2956,"children":2957},{"style":409},[2958],{"type":59,"value":412},{"type":53,"tag":397,"props":2960,"children":2961},{"style":409},[2962],{"type":59,"value":2904},{"type":53,"tag":397,"props":2964,"children":2965},{"style":409},[2966],{"type":59,"value":762},{"type":53,"tag":397,"props":2968,"children":2969},{"style":409},[2970],{"type":59,"value":2971}," --force",{"type":53,"tag":397,"props":2973,"children":2974},{"style":468},[2975],{"type":59,"value":2976},"   # skip confirmation prompt\n",{"type":53,"tag":525,"props":2978,"children":2980},{"id":2979},"flags-4",[2981],{"type":59,"value":530},{"type":53,"tag":532,"props":2983,"children":2984},{},[2985,3007],{"type":53,"tag":536,"props":2986,"children":2987},{},[2988],{"type":53,"tag":540,"props":2989,"children":2990},{},[2991,2995,2999,3003],{"type":53,"tag":544,"props":2992,"children":2993},{},[2994],{"type":59,"value":548},{"type":53,"tag":544,"props":2996,"children":2997},{},[2998],{"type":59,"value":553},{"type":53,"tag":544,"props":3000,"children":3001},{},[3002],{"type":59,"value":558},{"type":53,"tag":544,"props":3004,"children":3005},{},[3006],{"type":59,"value":563},{"type":53,"tag":565,"props":3008,"children":3009},{},[3010,3033,3056,3079],{"type":53,"tag":540,"props":3011,"children":3012},{},[3013,3021,3025,3029],{"type":53,"tag":572,"props":3014,"children":3015},{},[3016],{"type":53,"tag":74,"props":3017,"children":3019},{"className":3018},[],[3020],{"type":59,"value":855},{"type":53,"tag":572,"props":3022,"children":3023},{},[3024],{"type":59,"value":585},{"type":53,"tag":572,"props":3026,"children":3027},{},[3028],{"type":59,"value":864},{"type":53,"tag":572,"props":3030,"children":3031},{},[3032],{"type":59,"value":869},{"type":53,"tag":540,"props":3034,"children":3035},{},[3036,3044,3048,3052],{"type":53,"tag":572,"props":3037,"children":3038},{},[3039],{"type":53,"tag":74,"props":3040,"children":3042},{"className":3041},[],[3043],{"type":59,"value":580},{"type":53,"tag":572,"props":3045,"children":3046},{},[3047],{"type":59,"value":585},{"type":53,"tag":572,"props":3049,"children":3050},{},[3051],{"type":59,"value":590},{"type":53,"tag":572,"props":3053,"children":3054},{},[3055],{"type":59,"value":893},{"type":53,"tag":540,"props":3057,"children":3058},{},[3059,3067,3071,3075],{"type":53,"tag":572,"props":3060,"children":3061},{},[3062],{"type":53,"tag":74,"props":3063,"children":3065},{"className":3064},[],[3066],{"type":59,"value":87},{"type":53,"tag":572,"props":3068,"children":3069},{},[3070],{"type":59,"value":585},{"type":53,"tag":572,"props":3072,"children":3073},{},[3074],{"type":59,"value":590},{"type":53,"tag":572,"props":3076,"children":3077},{},[3078],{"type":59,"value":917},{"type":53,"tag":540,"props":3080,"children":3081},{},[3082,3091,3095,3099],{"type":53,"tag":572,"props":3083,"children":3084},{},[3085],{"type":53,"tag":74,"props":3086,"children":3088},{"className":3087},[],[3089],{"type":59,"value":3090},"--force, -f",{"type":53,"tag":572,"props":3092,"children":3093},{},[3094],{"type":59,"value":1526},{"type":53,"tag":572,"props":3096,"children":3097},{},[3098],{"type":59,"value":1531},{"type":53,"tag":572,"props":3100,"children":3101},{},[3102],{"type":59,"value":3103},"Skip confirmation prompt",{"type":53,"tag":136,"props":3105,"children":3107},{"id":3106},"annotate-runs-ax-experiments-annotate-runs",[3108,3110],{"type":59,"value":3109},"Annotate Runs: ",{"type":53,"tag":74,"props":3111,"children":3113},{"className":3112},[],[3114],{"type":59,"value":2847},{"type":53,"tag":66,"props":3116,"children":3117},{},[3118,3123,3125,3130,3132,3137],{"type":53,"tag":70,"props":3119,"children":3120},{},[3121],{"type":59,"value":3122},"This is the required step to attach evaluation scores\u002Flabels to an experiment and make them show up in the UI.",{"type":59,"value":3124}," Evaluations cannot be attached through ",{"type":53,"tag":74,"props":3126,"children":3128},{"className":3127},[],[3129],{"type":59,"value":2806},{"type":59,"value":3131},"; see the warning under Create Experiment. You write them here, after the experiment exists. Upsert semantics — resubmitting the same annotation ",{"type":53,"tag":74,"props":3133,"children":3135},{"className":3134},[],[3136],{"type":59,"value":102},{"type":59,"value":3138}," for the same run overwrites the previous value. Up to 1000 runs per request; unmatched record IDs are silently ignored.",{"type":53,"tag":386,"props":3140,"children":3142},{"className":388,"code":3141,"language":390,"meta":391,"style":391},"ax experiments annotate-runs NAME_OR_ID --file annotations.json --dataset DATASET_NAME --space SPACE\nax experiments annotate-runs NAME_OR_ID --file annotations.csv --dataset DATASET_NAME --space SPACE\n",[3143],{"type":53,"tag":74,"props":3144,"children":3145},{"__ignoreMap":391},[3146,3191],{"type":53,"tag":397,"props":3147,"children":3148},{"class":399,"line":400},[3149,3153,3157,3162,3166,3170,3175,3179,3183,3187],{"type":53,"tag":397,"props":3150,"children":3151},{"style":404},[3152],{"type":59,"value":224},{"type":53,"tag":397,"props":3154,"children":3155},{"style":409},[3156],{"type":59,"value":412},{"type":53,"tag":397,"props":3158,"children":3159},{"style":409},[3160],{"type":59,"value":3161}," annotate-runs",{"type":53,"tag":397,"props":3163,"children":3164},{"style":409},[3165],{"type":59,"value":762},{"type":53,"tag":397,"props":3167,"children":3168},{"style":409},[3169],{"type":59,"value":2195},{"type":53,"tag":397,"props":3171,"children":3172},{"style":409},[3173],{"type":59,"value":3174}," annotations.json",{"type":53,"tag":397,"props":3176,"children":3177},{"style":409},[3178],{"type":59,"value":439},{"type":53,"tag":397,"props":3180,"children":3181},{"style":409},[3182],{"type":59,"value":444},{"type":53,"tag":397,"props":3184,"children":3185},{"style":409},[3186],{"type":59,"value":449},{"type":53,"tag":397,"props":3188,"children":3189},{"style":409},[3190],{"type":59,"value":1197},{"type":53,"tag":397,"props":3192,"children":3193},{"class":399,"line":420},[3194,3198,3202,3206,3210,3214,3219,3223,3227,3231],{"type":53,"tag":397,"props":3195,"children":3196},{"style":404},[3197],{"type":59,"value":224},{"type":53,"tag":397,"props":3199,"children":3200},{"style":409},[3201],{"type":59,"value":412},{"type":53,"tag":397,"props":3203,"children":3204},{"style":409},[3205],{"type":59,"value":3161},{"type":53,"tag":397,"props":3207,"children":3208},{"style":409},[3209],{"type":59,"value":762},{"type":53,"tag":397,"props":3211,"children":3212},{"style":409},[3213],{"type":59,"value":2195},{"type":53,"tag":397,"props":3215,"children":3216},{"style":409},[3217],{"type":59,"value":3218}," annotations.csv",{"type":53,"tag":397,"props":3220,"children":3221},{"style":409},[3222],{"type":59,"value":439},{"type":53,"tag":397,"props":3224,"children":3225},{"style":409},[3226],{"type":59,"value":444},{"type":53,"tag":397,"props":3228,"children":3229},{"style":409},[3230],{"type":59,"value":449},{"type":53,"tag":397,"props":3232,"children":3233},{"style":409},[3234],{"type":59,"value":1197},{"type":53,"tag":525,"props":3236,"children":3238},{"id":3237},"annotation-file-schema",[3239],{"type":59,"value":3240},"Annotation file schema",{"type":53,"tag":66,"props":3242,"children":3243},{},[3244],{"type":59,"value":3245},"A JSON array; each item annotates one run:",{"type":53,"tag":386,"props":3247,"children":3249},{"className":1671,"code":3248,"language":1673,"meta":391,"style":391},"[\n  {\n    \"record_id\": \"run_001\",\n    \"values\": [\n      { \"name\": \"correctness\", \"label\": \"correct\", \"score\": 1.0 },\n      { \"name\": \"relevance\", \"score\": 0.95, \"text\": \"Directly answers the question\" }\n    ]\n  }\n]\n",[3250],{"type":53,"tag":74,"props":3251,"children":3252},{"__ignoreMap":391},[3253,3260,3267,3303,3328,3424,3519,3527,3534],{"type":53,"tag":397,"props":3254,"children":3255},{"class":399,"line":400},[3256],{"type":53,"tag":397,"props":3257,"children":3258},{"style":1385},[3259],{"type":59,"value":1685},{"type":53,"tag":397,"props":3261,"children":3262},{"class":399,"line":420},[3263],{"type":53,"tag":397,"props":3264,"children":3265},{"style":1385},[3266],{"type":59,"value":1693},{"type":53,"tag":397,"props":3268,"children":3269},{"class":399,"line":474},[3270,3274,3279,3283,3287,3291,3295,3299],{"type":53,"tag":397,"props":3271,"children":3272},{"style":1385},[3273],{"type":59,"value":1701},{"type":53,"tag":397,"props":3275,"children":3276},{"style":1704},[3277],{"type":59,"value":3278},"record_id",{"type":53,"tag":397,"props":3280,"children":3281},{"style":1385},[3282],{"type":59,"value":1711},{"type":53,"tag":397,"props":3284,"children":3285},{"style":1385},[3286],{"type":59,"value":1716},{"type":53,"tag":397,"props":3288,"children":3289},{"style":1385},[3290],{"type":59,"value":1721},{"type":53,"tag":397,"props":3292,"children":3293},{"style":409},[3294],{"type":59,"value":1726},{"type":53,"tag":397,"props":3296,"children":3297},{"style":1385},[3298],{"type":59,"value":1711},{"type":53,"tag":397,"props":3300,"children":3301},{"style":1385},[3302],{"type":59,"value":1735},{"type":53,"tag":397,"props":3304,"children":3305},{"class":399,"line":500},[3306,3310,3315,3319,3323],{"type":53,"tag":397,"props":3307,"children":3308},{"style":1385},[3309],{"type":59,"value":1701},{"type":53,"tag":397,"props":3311,"children":3312},{"style":1704},[3313],{"type":59,"value":3314},"values",{"type":53,"tag":397,"props":3316,"children":3317},{"style":1385},[3318],{"type":59,"value":1711},{"type":53,"tag":397,"props":3320,"children":3321},{"style":1385},[3322],{"type":59,"value":1716},{"type":53,"tag":397,"props":3324,"children":3325},{"style":1385},[3326],{"type":59,"value":3327}," [\n",{"type":53,"tag":397,"props":3329,"children":3330},{"class":399,"line":27},[3331,3336,3340,3344,3348,3352,3356,3360,3364,3368,3372,3376,3380,3384,3388,3392,3396,3400,3404,3408,3412,3416,3420],{"type":53,"tag":397,"props":3332,"children":3333},{"style":1385},[3334],{"type":59,"value":3335},"      {",{"type":53,"tag":397,"props":3337,"children":3338},{"style":1385},[3339],{"type":59,"value":1721},{"type":53,"tag":397,"props":3341,"children":3342},{"style":404},[3343],{"type":59,"value":102},{"type":53,"tag":397,"props":3345,"children":3346},{"style":1385},[3347],{"type":59,"value":1711},{"type":53,"tag":397,"props":3349,"children":3350},{"style":1385},[3351],{"type":59,"value":1716},{"type":53,"tag":397,"props":3353,"children":3354},{"style":1385},[3355],{"type":59,"value":1721},{"type":53,"tag":397,"props":3357,"children":3358},{"style":409},[3359],{"type":59,"value":192},{"type":53,"tag":397,"props":3361,"children":3362},{"style":1385},[3363],{"type":59,"value":1711},{"type":53,"tag":397,"props":3365,"children":3366},{"style":1385},[3367],{"type":59,"value":1894},{"type":53,"tag":397,"props":3369,"children":3370},{"style":1385},[3371],{"type":59,"value":1721},{"type":53,"tag":397,"props":3373,"children":3374},{"style":404},[3375],{"type":59,"value":1868},{"type":53,"tag":397,"props":3377,"children":3378},{"style":1385},[3379],{"type":59,"value":1711},{"type":53,"tag":397,"props":3381,"children":3382},{"style":1385},[3383],{"type":59,"value":1716},{"type":53,"tag":397,"props":3385,"children":3386},{"style":1385},[3387],{"type":59,"value":1721},{"type":53,"tag":397,"props":3389,"children":3390},{"style":409},[3391],{"type":59,"value":1885},{"type":53,"tag":397,"props":3393,"children":3394},{"style":1385},[3395],{"type":59,"value":1711},{"type":53,"tag":397,"props":3397,"children":3398},{"style":1385},[3399],{"type":59,"value":1894},{"type":53,"tag":397,"props":3401,"children":3402},{"style":1385},[3403],{"type":59,"value":1721},{"type":53,"tag":397,"props":3405,"children":3406},{"style":404},[3407],{"type":59,"value":1903},{"type":53,"tag":397,"props":3409,"children":3410},{"style":1385},[3411],{"type":59,"value":1711},{"type":53,"tag":397,"props":3413,"children":3414},{"style":1385},[3415],{"type":59,"value":1716},{"type":53,"tag":397,"props":3417,"children":3418},{"style":462},[3419],{"type":59,"value":1916},{"type":53,"tag":397,"props":3421,"children":3422},{"style":1385},[3423],{"type":59,"value":1921},{"type":53,"tag":397,"props":3425,"children":3426},{"class":399,"line":1257},[3427,3431,3435,3439,3443,3447,3451,3455,3459,3463,3467,3471,3475,3479,3483,3487,3491,3495,3499,3503,3507,3511,3515],{"type":53,"tag":397,"props":3428,"children":3429},{"style":1385},[3430],{"type":59,"value":3335},{"type":53,"tag":397,"props":3432,"children":3433},{"style":1385},[3434],{"type":59,"value":1721},{"type":53,"tag":397,"props":3436,"children":3437},{"style":404},[3438],{"type":59,"value":102},{"type":53,"tag":397,"props":3440,"children":3441},{"style":1385},[3442],{"type":59,"value":1711},{"type":53,"tag":397,"props":3444,"children":3445},{"style":1385},[3446],{"type":59,"value":1716},{"type":53,"tag":397,"props":3448,"children":3449},{"style":1385},[3450],{"type":59,"value":1721},{"type":53,"tag":397,"props":3452,"children":3453},{"style":409},[3454],{"type":59,"value":200},{"type":53,"tag":397,"props":3456,"children":3457},{"style":1385},[3458],{"type":59,"value":1711},{"type":53,"tag":397,"props":3460,"children":3461},{"style":1385},[3462],{"type":59,"value":1894},{"type":53,"tag":397,"props":3464,"children":3465},{"style":1385},[3466],{"type":59,"value":1721},{"type":53,"tag":397,"props":3468,"children":3469},{"style":404},[3470],{"type":59,"value":1903},{"type":53,"tag":397,"props":3472,"children":3473},{"style":1385},[3474],{"type":59,"value":1711},{"type":53,"tag":397,"props":3476,"children":3477},{"style":1385},[3478],{"type":59,"value":1716},{"type":53,"tag":397,"props":3480,"children":3481},{"style":462},[3482],{"type":59,"value":1965},{"type":53,"tag":397,"props":3484,"children":3485},{"style":1385},[3486],{"type":59,"value":1894},{"type":53,"tag":397,"props":3488,"children":3489},{"style":1385},[3490],{"type":59,"value":1721},{"type":53,"tag":397,"props":3492,"children":3493},{"style":404},[3494],{"type":59,"value":59},{"type":53,"tag":397,"props":3496,"children":3497},{"style":1385},[3498],{"type":59,"value":1711},{"type":53,"tag":397,"props":3500,"children":3501},{"style":1385},[3502],{"type":59,"value":1716},{"type":53,"tag":397,"props":3504,"children":3505},{"style":1385},[3506],{"type":59,"value":1721},{"type":53,"tag":397,"props":3508,"children":3509},{"style":409},[3510],{"type":59,"value":1995},{"type":53,"tag":397,"props":3512,"children":3513},{"style":1385},[3514],{"type":59,"value":1711},{"type":53,"tag":397,"props":3516,"children":3517},{"style":1385},[3518],{"type":59,"value":2004},{"type":53,"tag":397,"props":3520,"children":3521},{"class":399,"line":1303},[3522],{"type":53,"tag":397,"props":3523,"children":3524},{"style":1385},[3525],{"type":59,"value":3526},"    ]\n",{"type":53,"tag":397,"props":3528,"children":3529},{"class":399,"line":1344},[3530],{"type":53,"tag":397,"props":3531,"children":3532},{"style":1385},[3533],{"type":59,"value":2107},{"type":53,"tag":397,"props":3535,"children":3536},{"class":399,"line":2007},[3537],{"type":53,"tag":397,"props":3538,"children":3539},{"style":1385},[3540],{"type":59,"value":2116},{"type":53,"tag":532,"props":3542,"children":3543},{},[3544,3566],{"type":53,"tag":536,"props":3545,"children":3546},{},[3547],{"type":53,"tag":540,"props":3548,"children":3549},{},[3550,3554,3558,3562],{"type":53,"tag":544,"props":3551,"children":3552},{},[3553],{"type":59,"value":961},{"type":53,"tag":544,"props":3555,"children":3556},{},[3557],{"type":59,"value":553},{"type":53,"tag":544,"props":3559,"children":3560},{},[3561],{"type":59,"value":2284},{"type":53,"tag":544,"props":3563,"children":3564},{},[3565],{"type":59,"value":563},{"type":53,"tag":565,"props":3567,"children":3568},{},[3569,3624,3674,3706,3746,3784],{"type":53,"tag":540,"props":3570,"children":3571},{},[3572,3580,3584,3588],{"type":53,"tag":572,"props":3573,"children":3574},{},[3575],{"type":53,"tag":74,"props":3576,"children":3578},{"className":3577},[],[3579],{"type":59,"value":3278},{"type":53,"tag":572,"props":3581,"children":3582},{},[3583],{"type":59,"value":585},{"type":53,"tag":572,"props":3585,"children":3586},{},[3587],{"type":59,"value":2311},{"type":53,"tag":572,"props":3589,"children":3590},{},[3591,3593,3598,3600,3605,3606,3611,3613,3617,3619],{"type":59,"value":3592},"The ",{"type":53,"tag":70,"props":3594,"children":3595},{},[3596],{"type":59,"value":3597},"experiment run ID",{"type":59,"value":3599}," (the run's ",{"type":53,"tag":74,"props":3601,"children":3603},{"className":3602},[],[3604],{"type":59,"value":984},{"type":59,"value":2689},{"type":53,"tag":74,"props":3607,"children":3609},{"className":3608},[],[3610],{"type":59,"value":1131},{"type":59,"value":3612},") — ",{"type":53,"tag":70,"props":3614,"children":3615},{},[3616],{"type":59,"value":2770},{"type":59,"value":3618}," the ",{"type":53,"tag":74,"props":3620,"children":3622},{"className":3621},[],[3623],{"type":59,"value":1747},{"type":53,"tag":540,"props":3625,"children":3626},{},[3627,3635,3640,3644],{"type":53,"tag":572,"props":3628,"children":3629},{},[3630],{"type":53,"tag":74,"props":3631,"children":3633},{"className":3632},[],[3634],{"type":59,"value":3314},{"type":53,"tag":572,"props":3636,"children":3637},{},[3638],{"type":59,"value":3639},"array",{"type":53,"tag":572,"props":3641,"children":3642},{},[3643],{"type":59,"value":2311},{"type":53,"tag":572,"props":3645,"children":3646},{},[3647,3649,3654,3656,3661,3662,3667,3669],{"type":59,"value":3648},"One or more annotation dicts, each with a ",{"type":53,"tag":74,"props":3650,"children":3652},{"className":3651},[],[3653],{"type":59,"value":102},{"type":59,"value":3655}," plus at least one of ",{"type":53,"tag":74,"props":3657,"children":3659},{"className":3658},[],[3660],{"type":59,"value":1903},{"type":59,"value":194},{"type":53,"tag":74,"props":3663,"children":3665},{"className":3664},[],[3666],{"type":59,"value":1868},{"type":59,"value":3668},", or ",{"type":53,"tag":74,"props":3670,"children":3672},{"className":3671},[],[3673],{"type":59,"value":59},{"type":53,"tag":540,"props":3675,"children":3676},{},[3677,3686,3690,3694],{"type":53,"tag":572,"props":3678,"children":3679},{},[3680],{"type":53,"tag":74,"props":3681,"children":3683},{"className":3682},[],[3684],{"type":59,"value":3685},"values[].name",{"type":53,"tag":572,"props":3687,"children":3688},{},[3689],{"type":59,"value":585},{"type":53,"tag":572,"props":3691,"children":3692},{},[3693],{"type":59,"value":2311},{"type":53,"tag":572,"props":3695,"children":3696},{},[3697,3699,3704],{"type":59,"value":3698},"Annotation\u002Fevaluation name (e.g., ",{"type":53,"tag":74,"props":3700,"children":3702},{"className":3701},[],[3703],{"type":59,"value":192},{"type":59,"value":3705},") — becomes the score column in the UI",{"type":53,"tag":540,"props":3707,"children":3708},{},[3709,3718,3723,3727],{"type":53,"tag":572,"props":3710,"children":3711},{},[3712],{"type":53,"tag":74,"props":3713,"children":3715},{"className":3714},[],[3716],{"type":59,"value":3717},"values[].score",{"type":53,"tag":572,"props":3719,"children":3720},{},[3721],{"type":59,"value":3722},"number",{"type":53,"tag":572,"props":3724,"children":3725},{},[3726],{"type":59,"value":2360},{"type":53,"tag":572,"props":3728,"children":3729},{},[3730,3732,3738,3740,3745],{"type":59,"value":3731},"Numeric score (e.g., ",{"type":53,"tag":74,"props":3733,"children":3735},{"className":3734},[],[3736],{"type":59,"value":3737},"0.0",{"type":59,"value":3739},"–",{"type":53,"tag":74,"props":3741,"children":3743},{"className":3742},[],[3744],{"type":59,"value":47},{"type":59,"value":1603},{"type":53,"tag":540,"props":3747,"children":3748},{},[3749,3758,3762,3766],{"type":53,"tag":572,"props":3750,"children":3751},{},[3752],{"type":53,"tag":74,"props":3753,"children":3755},{"className":3754},[],[3756],{"type":59,"value":3757},"values[].label",{"type":53,"tag":572,"props":3759,"children":3760},{},[3761],{"type":59,"value":585},{"type":53,"tag":572,"props":3763,"children":3764},{},[3765],{"type":59,"value":2360},{"type":53,"tag":572,"props":3767,"children":3768},{},[3769,3771,3776,3777,3783],{"type":59,"value":3770},"Categorical label (e.g., ",{"type":53,"tag":74,"props":3772,"children":3774},{"className":3773},[],[3775],{"type":59,"value":1885},{"type":59,"value":194},{"type":53,"tag":74,"props":3778,"children":3780},{"className":3779},[],[3781],{"type":59,"value":3782},"incorrect",{"type":59,"value":1603},{"type":53,"tag":540,"props":3785,"children":3786},{},[3787,3796,3800,3804],{"type":53,"tag":572,"props":3788,"children":3789},{},[3790],{"type":53,"tag":74,"props":3791,"children":3793},{"className":3792},[],[3794],{"type":59,"value":3795},"values[].text",{"type":53,"tag":572,"props":3797,"children":3798},{},[3799],{"type":59,"value":585},{"type":53,"tag":572,"props":3801,"children":3802},{},[3803],{"type":59,"value":2360},{"type":53,"tag":572,"props":3805,"children":3806},{},[3807],{"type":59,"value":3808},"Freeform explanation",{"type":53,"tag":62,"props":3810,"children":3811},{},[3812],{"type":53,"tag":66,"props":3813,"children":3814},{},[3815,3820,3822,3827,3829,3834,3836,3841,3843,3849,3851,3856,3858,3864],{"type":53,"tag":74,"props":3816,"children":3818},{"className":3817},[],[3819],{"type":59,"value":3278},{"type":59,"value":3821}," keys on the ",{"type":53,"tag":70,"props":3823,"children":3824},{},[3825],{"type":59,"value":3826},"run",{"type":59,"value":3828}," id, which only exists after ",{"type":53,"tag":74,"props":3830,"children":3832},{"className":3831},[],[3833],{"type":59,"value":2806},{"type":59,"value":3835},". So the order is always: ",{"type":53,"tag":74,"props":3837,"children":3839},{"className":3838},[],[3840],{"type":59,"value":2806},{"type":59,"value":3842}," → ",{"type":53,"tag":74,"props":3844,"children":3846},{"className":3845},[],[3847],{"type":59,"value":3848},"export",{"type":59,"value":3850}," (to read each run's ",{"type":53,"tag":74,"props":3852,"children":3854},{"className":3853},[],[3855],{"type":59,"value":984},{"type":59,"value":3857},") → build annotations → ",{"type":53,"tag":74,"props":3859,"children":3861},{"className":3860},[],[3862],{"type":59,"value":3863},"annotate-runs",{"type":59,"value":134},{"type":53,"tag":525,"props":3866,"children":3868},{"id":3867},"flags-5",[3869],{"type":59,"value":530},{"type":53,"tag":532,"props":3871,"children":3872},{},[3873,3895],{"type":53,"tag":536,"props":3874,"children":3875},{},[3876],{"type":53,"tag":540,"props":3877,"children":3878},{},[3879,3883,3887,3891],{"type":53,"tag":544,"props":3880,"children":3881},{},[3882],{"type":59,"value":548},{"type":53,"tag":544,"props":3884,"children":3885},{},[3886],{"type":59,"value":553},{"type":53,"tag":544,"props":3888,"children":3889},{},[3890],{"type":59,"value":2284},{"type":53,"tag":544,"props":3892,"children":3893},{},[3894],{"type":59,"value":563},{"type":53,"tag":565,"props":3896,"children":3897},{},[3898,3921,3953,3977],{"type":53,"tag":540,"props":3899,"children":3900},{},[3901,3909,3913,3917],{"type":53,"tag":572,"props":3902,"children":3903},{},[3904],{"type":53,"tag":74,"props":3905,"children":3907},{"className":3906},[],[3908],{"type":59,"value":855},{"type":53,"tag":572,"props":3910,"children":3911},{},[3912],{"type":59,"value":585},{"type":53,"tag":572,"props":3914,"children":3915},{},[3916],{"type":59,"value":2311},{"type":53,"tag":572,"props":3918,"children":3919},{},[3920],{"type":59,"value":869},{"type":53,"tag":540,"props":3922,"children":3923},{},[3924,3932,3936,3940],{"type":53,"tag":572,"props":3925,"children":3926},{},[3927],{"type":53,"tag":74,"props":3928,"children":3930},{"className":3929},[],[3931],{"type":59,"value":2376},{"type":53,"tag":572,"props":3933,"children":3934},{},[3935],{"type":59,"value":2381},{"type":53,"tag":572,"props":3937,"children":3938},{},[3939],{"type":59,"value":2311},{"type":53,"tag":572,"props":3941,"children":3942},{},[3943,3945,3951],{"type":59,"value":3944},"Annotation file: JSON, JSONL, CSV, or Parquet (use ",{"type":53,"tag":74,"props":3946,"children":3948},{"className":3947},[],[3949],{"type":59,"value":3950},"-",{"type":59,"value":3952}," for stdin)",{"type":53,"tag":540,"props":3954,"children":3955},{},[3956,3964,3968,3972],{"type":53,"tag":572,"props":3957,"children":3958},{},[3959],{"type":53,"tag":74,"props":3960,"children":3962},{"className":3961},[],[3963],{"type":59,"value":580},{"type":53,"tag":572,"props":3965,"children":3966},{},[3967],{"type":59,"value":585},{"type":53,"tag":572,"props":3969,"children":3970},{},[3971],{"type":59,"value":2311},{"type":53,"tag":572,"props":3973,"children":3974},{},[3975],{"type":59,"value":3976},"Dataset name or ID (required when using experiment name instead of ID)",{"type":53,"tag":540,"props":3978,"children":3979},{},[3980,3988,3992,3996],{"type":53,"tag":572,"props":3981,"children":3982},{},[3983],{"type":53,"tag":74,"props":3984,"children":3986},{"className":3985},[],[3987],{"type":59,"value":87},{"type":53,"tag":572,"props":3989,"children":3990},{},[3991],{"type":59,"value":585},{"type":53,"tag":572,"props":3993,"children":3994},{},[3995],{"type":59,"value":2360},{"type":53,"tag":572,"props":3997,"children":3998},{},[3999],{"type":59,"value":4000},"Space name or ID",{"type":53,"tag":136,"props":4002,"children":4004},{"id":4003},"experiment-run-schema",[4005],{"type":59,"value":4006},"Experiment Run Schema",{"type":53,"tag":66,"props":4008,"children":4009},{},[4010,4012,4037,4039,4044,4046,4051,4053,4059,4061,4066,4068,4073],{"type":59,"value":4011},"Each run corresponds to one dataset example. ",{"type":53,"tag":70,"props":4013,"children":4014},{},[4015,4017,4022,4024,4029,4030,4035],{"type":59,"value":4016},"On ",{"type":53,"tag":74,"props":4018,"children":4020},{"className":4019},[],[4021],{"type":59,"value":2806},{"type":59,"value":4023},", only ",{"type":53,"tag":74,"props":4025,"children":4027},{"className":4026},[],[4028],{"type":59,"value":1747},{"type":59,"value":2815},{"type":53,"tag":74,"props":4031,"children":4033},{"className":4032},[],[4034],{"type":59,"value":1784},{"type":59,"value":4036}," are consumed",{"type":59,"value":4038}," — ",{"type":53,"tag":74,"props":4040,"children":4042},{"className":4041},[],[4043],{"type":59,"value":1821},{"type":59,"value":4045}," shown here is the shape ",{"type":53,"tag":74,"props":4047,"children":4049},{"className":4048},[],[4050],{"type":59,"value":3848},{"type":59,"value":4052}," returns ",{"type":53,"tag":4054,"props":4055,"children":4056},"em",{},[4057],{"type":59,"value":4058},"after",{"type":59,"value":4060}," you attach scores via ",{"type":53,"tag":74,"props":4062,"children":4064},{"className":4063},[],[4065],{"type":59,"value":3863},{"type":59,"value":4067},"; it is not an input to ",{"type":53,"tag":74,"props":4069,"children":4071},{"className":4070},[],[4072],{"type":59,"value":2806},{"type":59,"value":134},{"type":53,"tag":386,"props":4075,"children":4077},{"className":1671,"code":4076,"language":1673,"meta":391,"style":391},"{\n  \"example_id\": \"required on create -- the dataset example's top-level id\",\n  \"output\": \"required on create -- the model\u002Fsystem output for this example\",\n  \"evaluations\": {\n    \"metric_name\": {\n      \"label\": \"optional string label (e.g., 'correct', 'incorrect')\",\n      \"score\": \"optional numeric score (e.g., 0.95)\",\n      \"explanation\": \"optional freeform text\"\n    }\n  },\n  \"metadata\": {\n    \"model\": \"gpt-4o\",\n    \"temperature\": 0.7,\n    \"latency_ms\": 1234\n  }\n}\n",[4078],{"type":53,"tag":74,"props":4079,"children":4080},{"__ignoreMap":391},[4081,4089,4126,4162,4185,4209,4245,4281,4314,4322,4330,4353,4388,4418,4443,4451],{"type":53,"tag":397,"props":4082,"children":4083},{"class":399,"line":400},[4084],{"type":53,"tag":397,"props":4085,"children":4086},{"style":1385},[4087],{"type":59,"value":4088},"{\n",{"type":53,"tag":397,"props":4090,"children":4091},{"class":399,"line":420},[4092,4097,4101,4105,4109,4113,4118,4122],{"type":53,"tag":397,"props":4093,"children":4094},{"style":1385},[4095],{"type":59,"value":4096},"  \"",{"type":53,"tag":397,"props":4098,"children":4099},{"style":1704},[4100],{"type":59,"value":1747},{"type":53,"tag":397,"props":4102,"children":4103},{"style":1385},[4104],{"type":59,"value":1711},{"type":53,"tag":397,"props":4106,"children":4107},{"style":1385},[4108],{"type":59,"value":1716},{"type":53,"tag":397,"props":4110,"children":4111},{"style":1385},[4112],{"type":59,"value":1721},{"type":53,"tag":397,"props":4114,"children":4115},{"style":409},[4116],{"type":59,"value":4117},"required on create -- the dataset example's top-level id",{"type":53,"tag":397,"props":4119,"children":4120},{"style":1385},[4121],{"type":59,"value":1711},{"type":53,"tag":397,"props":4123,"children":4124},{"style":1385},[4125],{"type":59,"value":1735},{"type":53,"tag":397,"props":4127,"children":4128},{"class":399,"line":474},[4129,4133,4137,4141,4145,4149,4154,4158],{"type":53,"tag":397,"props":4130,"children":4131},{"style":1385},[4132],{"type":59,"value":4096},{"type":53,"tag":397,"props":4134,"children":4135},{"style":1704},[4136],{"type":59,"value":1784},{"type":53,"tag":397,"props":4138,"children":4139},{"style":1385},[4140],{"type":59,"value":1711},{"type":53,"tag":397,"props":4142,"children":4143},{"style":1385},[4144],{"type":59,"value":1716},{"type":53,"tag":397,"props":4146,"children":4147},{"style":1385},[4148],{"type":59,"value":1721},{"type":53,"tag":397,"props":4150,"children":4151},{"style":409},[4152],{"type":59,"value":4153},"required on create -- the model\u002Fsystem output for this example",{"type":53,"tag":397,"props":4155,"children":4156},{"style":1385},[4157],{"type":59,"value":1711},{"type":53,"tag":397,"props":4159,"children":4160},{"style":1385},[4161],{"type":59,"value":1735},{"type":53,"tag":397,"props":4163,"children":4164},{"class":399,"line":500},[4165,4169,4173,4177,4181],{"type":53,"tag":397,"props":4166,"children":4167},{"style":1385},[4168],{"type":59,"value":4096},{"type":53,"tag":397,"props":4170,"children":4171},{"style":1704},[4172],{"type":59,"value":1821},{"type":53,"tag":397,"props":4174,"children":4175},{"style":1385},[4176],{"type":59,"value":1711},{"type":53,"tag":397,"props":4178,"children":4179},{"style":1385},[4180],{"type":59,"value":1716},{"type":53,"tag":397,"props":4182,"children":4183},{"style":1385},[4184],{"type":59,"value":1834},{"type":53,"tag":397,"props":4186,"children":4187},{"class":399,"line":27},[4188,4192,4197,4201,4205],{"type":53,"tag":397,"props":4189,"children":4190},{"style":1385},[4191],{"type":59,"value":1701},{"type":53,"tag":397,"props":4193,"children":4194},{"style":404},[4195],{"type":59,"value":4196},"metric_name",{"type":53,"tag":397,"props":4198,"children":4199},{"style":1385},[4200],{"type":59,"value":1711},{"type":53,"tag":397,"props":4202,"children":4203},{"style":1385},[4204],{"type":59,"value":1716},{"type":53,"tag":397,"props":4206,"children":4207},{"style":1385},[4208],{"type":59,"value":1834},{"type":53,"tag":397,"props":4210,"children":4211},{"class":399,"line":1257},[4212,4216,4220,4224,4228,4232,4237,4241],{"type":53,"tag":397,"props":4213,"children":4214},{"style":1385},[4215],{"type":59,"value":1842},{"type":53,"tag":397,"props":4217,"children":4218},{"style":462},[4219],{"type":59,"value":1868},{"type":53,"tag":397,"props":4221,"children":4222},{"style":1385},[4223],{"type":59,"value":1711},{"type":53,"tag":397,"props":4225,"children":4226},{"style":1385},[4227],{"type":59,"value":1716},{"type":53,"tag":397,"props":4229,"children":4230},{"style":1385},[4231],{"type":59,"value":1721},{"type":53,"tag":397,"props":4233,"children":4234},{"style":409},[4235],{"type":59,"value":4236},"optional string label (e.g., 'correct', 'incorrect')",{"type":53,"tag":397,"props":4238,"children":4239},{"style":1385},[4240],{"type":59,"value":1711},{"type":53,"tag":397,"props":4242,"children":4243},{"style":1385},[4244],{"type":59,"value":1735},{"type":53,"tag":397,"props":4246,"children":4247},{"class":399,"line":1303},[4248,4252,4256,4260,4264,4268,4273,4277],{"type":53,"tag":397,"props":4249,"children":4250},{"style":1385},[4251],{"type":59,"value":1842},{"type":53,"tag":397,"props":4253,"children":4254},{"style":462},[4255],{"type":59,"value":1903},{"type":53,"tag":397,"props":4257,"children":4258},{"style":1385},[4259],{"type":59,"value":1711},{"type":53,"tag":397,"props":4261,"children":4262},{"style":1385},[4263],{"type":59,"value":1716},{"type":53,"tag":397,"props":4265,"children":4266},{"style":1385},[4267],{"type":59,"value":1721},{"type":53,"tag":397,"props":4269,"children":4270},{"style":409},[4271],{"type":59,"value":4272},"optional numeric score (e.g., 0.95)",{"type":53,"tag":397,"props":4274,"children":4275},{"style":1385},[4276],{"type":59,"value":1711},{"type":53,"tag":397,"props":4278,"children":4279},{"style":1385},[4280],{"type":59,"value":1735},{"type":53,"tag":397,"props":4282,"children":4283},{"class":399,"line":1344},[4284,4288,4292,4296,4300,4304,4309],{"type":53,"tag":397,"props":4285,"children":4286},{"style":1385},[4287],{"type":59,"value":1842},{"type":53,"tag":397,"props":4289,"children":4290},{"style":462},[4291],{"type":59,"value":1978},{"type":53,"tag":397,"props":4293,"children":4294},{"style":1385},[4295],{"type":59,"value":1711},{"type":53,"tag":397,"props":4297,"children":4298},{"style":1385},[4299],{"type":59,"value":1716},{"type":53,"tag":397,"props":4301,"children":4302},{"style":1385},[4303],{"type":59,"value":1721},{"type":53,"tag":397,"props":4305,"children":4306},{"style":409},[4307],{"type":59,"value":4308},"optional freeform text",{"type":53,"tag":397,"props":4310,"children":4311},{"style":1385},[4312],{"type":59,"value":4313},"\"\n",{"type":53,"tag":397,"props":4315,"children":4316},{"class":399,"line":2007},[4317],{"type":53,"tag":397,"props":4318,"children":4319},{"style":1385},[4320],{"type":59,"value":4321},"    }\n",{"type":53,"tag":397,"props":4323,"children":4324},{"class":399,"line":2016},[4325],{"type":53,"tag":397,"props":4326,"children":4327},{"style":1385},[4328],{"type":59,"value":4329},"  },\n",{"type":53,"tag":397,"props":4331,"children":4332},{"class":399,"line":2101},[4333,4337,4341,4345,4349],{"type":53,"tag":397,"props":4334,"children":4335},{"style":1385},[4336],{"type":59,"value":4096},{"type":53,"tag":397,"props":4338,"children":4339},{"style":1704},[4340],{"type":59,"value":2026},{"type":53,"tag":397,"props":4342,"children":4343},{"style":1385},[4344],{"type":59,"value":1711},{"type":53,"tag":397,"props":4346,"children":4347},{"style":1385},[4348],{"type":59,"value":1716},{"type":53,"tag":397,"props":4350,"children":4351},{"style":1385},[4352],{"type":59,"value":1834},{"type":53,"tag":397,"props":4354,"children":4355},{"class":399,"line":2110},[4356,4360,4364,4368,4372,4376,4380,4384],{"type":53,"tag":397,"props":4357,"children":4358},{"style":1385},[4359],{"type":59,"value":1701},{"type":53,"tag":397,"props":4361,"children":4362},{"style":404},[4363],{"type":59,"value":2047},{"type":53,"tag":397,"props":4365,"children":4366},{"style":1385},[4367],{"type":59,"value":1711},{"type":53,"tag":397,"props":4369,"children":4370},{"style":1385},[4371],{"type":59,"value":1716},{"type":53,"tag":397,"props":4373,"children":4374},{"style":1385},[4375],{"type":59,"value":1721},{"type":53,"tag":397,"props":4377,"children":4378},{"style":409},[4379],{"type":59,"value":2064},{"type":53,"tag":397,"props":4381,"children":4382},{"style":1385},[4383],{"type":59,"value":1711},{"type":53,"tag":397,"props":4385,"children":4386},{"style":1385},[4387],{"type":59,"value":1735},{"type":53,"tag":397,"props":4389,"children":4391},{"class":399,"line":4390},13,[4392,4396,4401,4405,4409,4414],{"type":53,"tag":397,"props":4393,"children":4394},{"style":1385},[4395],{"type":59,"value":1701},{"type":53,"tag":397,"props":4397,"children":4398},{"style":404},[4399],{"type":59,"value":4400},"temperature",{"type":53,"tag":397,"props":4402,"children":4403},{"style":1385},[4404],{"type":59,"value":1711},{"type":53,"tag":397,"props":4406,"children":4407},{"style":1385},[4408],{"type":59,"value":1716},{"type":53,"tag":397,"props":4410,"children":4411},{"style":462},[4412],{"type":59,"value":4413}," 0.7",{"type":53,"tag":397,"props":4415,"children":4416},{"style":1385},[4417],{"type":59,"value":1735},{"type":53,"tag":397,"props":4419,"children":4421},{"class":399,"line":4420},14,[4422,4426,4430,4434,4438],{"type":53,"tag":397,"props":4423,"children":4424},{"style":1385},[4425],{"type":59,"value":1701},{"type":53,"tag":397,"props":4427,"children":4428},{"style":404},[4429],{"type":59,"value":2081},{"type":53,"tag":397,"props":4431,"children":4432},{"style":1385},[4433],{"type":59,"value":1711},{"type":53,"tag":397,"props":4435,"children":4436},{"style":1385},[4437],{"type":59,"value":1716},{"type":53,"tag":397,"props":4439,"children":4440},{"style":462},[4441],{"type":59,"value":4442}," 1234\n",{"type":53,"tag":397,"props":4444,"children":4446},{"class":399,"line":4445},15,[4447],{"type":53,"tag":397,"props":4448,"children":4449},{"style":1385},[4450],{"type":59,"value":2107},{"type":53,"tag":397,"props":4452,"children":4454},{"class":399,"line":4453},16,[4455],{"type":53,"tag":397,"props":4456,"children":4457},{"style":1385},[4458],{"type":59,"value":4459},"}\n",{"type":53,"tag":525,"props":4461,"children":4463},{"id":4462},"evaluation-fields",[4464],{"type":59,"value":4465},"Evaluation fields",{"type":53,"tag":532,"props":4467,"children":4468},{},[4469,4491],{"type":53,"tag":536,"props":4470,"children":4471},{},[4472],{"type":53,"tag":540,"props":4473,"children":4474},{},[4475,4479,4483,4487],{"type":53,"tag":544,"props":4476,"children":4477},{},[4478],{"type":59,"value":961},{"type":53,"tag":544,"props":4480,"children":4481},{},[4482],{"type":59,"value":553},{"type":53,"tag":544,"props":4484,"children":4485},{},[4486],{"type":59,"value":2284},{"type":53,"tag":544,"props":4488,"children":4489},{},[4490],{"type":59,"value":563},{"type":53,"tag":565,"props":4492,"children":4493},{},[4494,4537,4561],{"type":53,"tag":540,"props":4495,"children":4496},{},[4497,4505,4509,4513],{"type":53,"tag":572,"props":4498,"children":4499},{},[4500],{"type":53,"tag":74,"props":4501,"children":4503},{"className":4502},[],[4504],{"type":59,"value":1868},{"type":53,"tag":572,"props":4506,"children":4507},{},[4508],{"type":59,"value":585},{"type":53,"tag":572,"props":4510,"children":4511},{},[4512],{"type":59,"value":2360},{"type":53,"tag":572,"props":4514,"children":4515},{},[4516,4518,4523,4524,4529,4530,4536],{"type":59,"value":4517},"Categorical classification (e.g., ",{"type":53,"tag":74,"props":4519,"children":4521},{"className":4520},[],[4522],{"type":59,"value":1885},{"type":59,"value":194},{"type":53,"tag":74,"props":4525,"children":4527},{"className":4526},[],[4528],{"type":59,"value":3782},{"type":59,"value":194},{"type":53,"tag":74,"props":4531,"children":4533},{"className":4532},[],[4534],{"type":59,"value":4535},"partial",{"type":59,"value":1603},{"type":53,"tag":540,"props":4538,"children":4539},{},[4540,4548,4552,4556],{"type":53,"tag":572,"props":4541,"children":4542},{},[4543],{"type":53,"tag":74,"props":4544,"children":4546},{"className":4545},[],[4547],{"type":59,"value":1903},{"type":53,"tag":572,"props":4549,"children":4550},{},[4551],{"type":59,"value":3722},{"type":53,"tag":572,"props":4553,"children":4554},{},[4555],{"type":59,"value":2360},{"type":53,"tag":572,"props":4557,"children":4558},{},[4559],{"type":59,"value":4560},"Numeric quality score (e.g., 0.0 - 1.0)",{"type":53,"tag":540,"props":4562,"children":4563},{},[4564,4572,4576,4580],{"type":53,"tag":572,"props":4565,"children":4566},{},[4567],{"type":53,"tag":74,"props":4568,"children":4570},{"className":4569},[],[4571],{"type":59,"value":1978},{"type":53,"tag":572,"props":4573,"children":4574},{},[4575],{"type":59,"value":585},{"type":53,"tag":572,"props":4577,"children":4578},{},[4579],{"type":59,"value":2360},{"type":53,"tag":572,"props":4581,"children":4582},{},[4583],{"type":59,"value":4584},"Freeform reasoning for the evaluation",{"type":53,"tag":66,"props":4586,"children":4587},{},[4588,4590,4595,4596,4601,4602,4607],{"type":59,"value":4589},"At least one of ",{"type":53,"tag":74,"props":4591,"children":4593},{"className":4592},[],[4594],{"type":59,"value":1868},{"type":59,"value":194},{"type":53,"tag":74,"props":4597,"children":4599},{"className":4598},[],[4600],{"type":59,"value":1903},{"type":59,"value":3668},{"type":53,"tag":74,"props":4603,"children":4605},{"className":4604},[],[4606],{"type":59,"value":1978},{"type":59,"value":4608}," should be present per evaluation.",{"type":53,"tag":136,"props":4610,"children":4612},{"id":4611},"workflows",[4613],{"type":59,"value":4614},"Workflows",{"type":53,"tag":525,"props":4616,"children":4618},{"id":4617},"run-an-experiment-against-a-dataset",[4619],{"type":59,"value":4620},"Run an experiment against a dataset",{"type":53,"tag":4622,"props":4623,"children":4624},"ol",{},[4625,4713,4752,5389,5480,5547,5588,5607,5822],{"type":53,"tag":147,"props":4626,"children":4627},{},[4628,4630],{"type":59,"value":4629},"Find or create a dataset:",{"type":53,"tag":386,"props":4631,"children":4633},{"className":388,"code":4632,"language":390,"meta":391,"style":391},"ax datasets list --space SPACE\nax datasets export DATASET_NAME --space SPACE --stdout | jq 'length'\n",[4634],{"type":53,"tag":74,"props":4635,"children":4636},{"__ignoreMap":391},[4637,4661],{"type":53,"tag":397,"props":4638,"children":4639},{"class":399,"line":400},[4640,4644,4649,4653,4657],{"type":53,"tag":397,"props":4641,"children":4642},{"style":404},[4643],{"type":59,"value":224},{"type":53,"tag":397,"props":4645,"children":4646},{"style":409},[4647],{"type":59,"value":4648}," datasets",{"type":53,"tag":397,"props":4650,"children":4651},{"style":409},[4652],{"type":59,"value":434},{"type":53,"tag":397,"props":4654,"children":4655},{"style":409},[4656],{"type":59,"value":449},{"type":53,"tag":397,"props":4658,"children":4659},{"style":409},[4660],{"type":59,"value":1197},{"type":53,"tag":397,"props":4662,"children":4663},{"class":399,"line":420},[4664,4668,4672,4676,4680,4684,4688,4692,4696,4700,4704,4709],{"type":53,"tag":397,"props":4665,"children":4666},{"style":404},[4667],{"type":59,"value":224},{"type":53,"tag":397,"props":4669,"children":4670},{"style":409},[4671],{"type":59,"value":4648},{"type":53,"tag":397,"props":4673,"children":4674},{"style":409},[4675],{"type":59,"value":1175},{"type":53,"tag":397,"props":4677,"children":4678},{"style":409},[4679],{"type":59,"value":444},{"type":53,"tag":397,"props":4681,"children":4682},{"style":409},[4683],{"type":59,"value":449},{"type":53,"tag":397,"props":4685,"children":4686},{"style":409},[4687],{"type":59,"value":454},{"type":53,"tag":397,"props":4689,"children":4690},{"style":409},[4691],{"type":59,"value":1382},{"type":53,"tag":397,"props":4693,"children":4694},{"style":1385},[4695],{"type":59,"value":1388},{"type":53,"tag":397,"props":4697,"children":4698},{"style":404},[4699],{"type":59,"value":1393},{"type":53,"tag":397,"props":4701,"children":4702},{"style":1385},[4703],{"type":59,"value":1398},{"type":53,"tag":397,"props":4705,"children":4706},{"style":409},[4707],{"type":59,"value":4708},"length",{"type":53,"tag":397,"props":4710,"children":4711},{"style":1385},[4712],{"type":59,"value":1408},{"type":53,"tag":147,"props":4714,"children":4715},{},[4716,4718],{"type":59,"value":4717},"Export the dataset examples:",{"type":53,"tag":386,"props":4719,"children":4721},{"className":388,"code":4720,"language":390,"meta":391,"style":391},"ax datasets export DATASET_NAME --space SPACE\n",[4722],{"type":53,"tag":74,"props":4723,"children":4724},{"__ignoreMap":391},[4725],{"type":53,"tag":397,"props":4726,"children":4727},{"class":399,"line":400},[4728,4732,4736,4740,4744,4748],{"type":53,"tag":397,"props":4729,"children":4730},{"style":404},[4731],{"type":59,"value":224},{"type":53,"tag":397,"props":4733,"children":4734},{"style":409},[4735],{"type":59,"value":4648},{"type":53,"tag":397,"props":4737,"children":4738},{"style":409},[4739],{"type":59,"value":1175},{"type":53,"tag":397,"props":4741,"children":4742},{"style":409},[4743],{"type":59,"value":444},{"type":53,"tag":397,"props":4745,"children":4746},{"style":409},[4747],{"type":59,"value":449},{"type":53,"tag":397,"props":4749,"children":4750},{"style":409},[4751],{"type":59,"value":1197},{"type":53,"tag":147,"props":4753,"children":4754},{},[4755,4757,4763,4765,4826,4830,4832,4838,4840,5356,5359,5364,5366,5372,5374,5380,5381,5387],{"type":59,"value":4756},"Call the real model API for each example and collect outputs. Use ",{"type":53,"tag":74,"props":4758,"children":4760},{"className":4759},[],[4761],{"type":59,"value":4762},"ax datasets export --stdout",{"type":59,"value":4764}," to pipe examples directly into an inference script:",{"type":53,"tag":386,"props":4766,"children":4768},{"className":388,"code":4767,"language":390,"meta":391,"style":391},"ax datasets export DATASET_NAME --space SPACE --stdout | python3 infer.py > runs.json\n",[4769],{"type":53,"tag":74,"props":4770,"children":4771},{"__ignoreMap":391},[4772],{"type":53,"tag":397,"props":4773,"children":4774},{"class":399,"line":400},[4775,4779,4783,4787,4791,4795,4799,4803,4807,4812,4817,4822],{"type":53,"tag":397,"props":4776,"children":4777},{"style":404},[4778],{"type":59,"value":224},{"type":53,"tag":397,"props":4780,"children":4781},{"style":409},[4782],{"type":59,"value":4648},{"type":53,"tag":397,"props":4784,"children":4785},{"style":409},[4786],{"type":59,"value":1175},{"type":53,"tag":397,"props":4788,"children":4789},{"style":409},[4790],{"type":59,"value":444},{"type":53,"tag":397,"props":4792,"children":4793},{"style":409},[4794],{"type":59,"value":449},{"type":53,"tag":397,"props":4796,"children":4797},{"style":409},[4798],{"type":59,"value":454},{"type":53,"tag":397,"props":4800,"children":4801},{"style":409},[4802],{"type":59,"value":1382},{"type":53,"tag":397,"props":4804,"children":4805},{"style":1385},[4806],{"type":59,"value":1388},{"type":53,"tag":397,"props":4808,"children":4809},{"style":404},[4810],{"type":59,"value":4811}," python3",{"type":53,"tag":397,"props":4813,"children":4814},{"style":409},[4815],{"type":59,"value":4816}," infer.py",{"type":53,"tag":397,"props":4818,"children":4819},{"style":1385},[4820],{"type":59,"value":4821}," >",{"type":53,"tag":397,"props":4823,"children":4824},{"style":409},[4825],{"type":59,"value":2200},{"type":53,"tag":4827,"props":4828,"children":4829},"br",{},[],{"type":59,"value":4831},"Write ",{"type":53,"tag":74,"props":4833,"children":4835},{"className":4834},[],[4836],{"type":59,"value":4837},"infer.py",{"type":59,"value":4839}," to read examples from stdin, call the target model, and write runs JSON to stdout. The script below is a template — first inspect the exported dataset JSON to find the correct input field name, then uncomment the provider block the user wants:",{"type":53,"tag":386,"props":4841,"children":4845},{"className":4842,"code":4843,"language":4844,"meta":391,"style":391},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","import json, sys, time\n\nexamples = json.load(sys.stdin)\nruns = []\n\nfor ex in examples:\n    # Inspect the exported JSON to find the right field (e.g. \"input\", \"question\", \"prompt\")\n    user_input = ex.get(\"input\") or ex.get(\"question\") or ex.get(\"prompt\") or str(ex)\n\n    start = time.time()\n\n    # === CALL THE REAL MODEL API HERE — never fabricate or simulate ===\n    # Uncomment and adapt the provider block the user requested:\n    #\n    # OpenAI (pip install openai  — uses OPENAI_API_KEY env var):\n    #   from openai import OpenAI\n    #   resp = OpenAI().chat.completions.create(\n    #       model=\"gpt-4o\",\n    #       messages=[{\"role\": \"user\", \"content\": user_input}]\n    #   )\n    #   output_text = resp.choices[0].message.content\n    #\n    # Anthropic (pip install anthropic  — uses ANTHROPIC_API_KEY env var):\n    #   import anthropic\n    #   resp = anthropic.Anthropic().messages.create(\n    #       model=\"claude-sonnet-4-6\", max_tokens=1024,\n    #       messages=[{\"role\": \"user\", \"content\": user_input}]\n    #   )\n    #   output_text = resp.content[0].text\n    #\n    # Google Gemini (pip install google-genai  — uses GOOGLE_API_KEY env var):\n    #   from google import genai\n    #   resp = genai.Client().models.generate_content(\n    #       model=\"gemini-2.5-pro\", contents=user_input\n    #   )\n    #   output_text = resp.text\n    #\n    # Custom \u002F OpenAI-compatible proxy (pip install openai — uses CUSTOM_BASE_URL + CUSTOM_API_KEY env vars):\n    # Use this for Azure OpenAI, NVIDIA NIM, local Ollama, or any OpenAI-compatible endpoint,\n    # including a test integration proxy. Matches the `custom` provider in `ax ai-integrations create`.\n    #   import os\n    #   from openai import OpenAI\n    #   resp = OpenAI(\n    #       base_url=os.environ[\"CUSTOM_BASE_URL\"],          # e.g. https:\u002F\u002Fmy-proxy.example.com\u002Fv1\n    #       api_key=os.environ.get(\"CUSTOM_API_KEY\", \"none\"),\n    #   ).chat.completions.create(\n    #       model=os.environ.get(\"CUSTOM_MODEL\", \"default\"),\n    #       messages=[{\"role\": \"user\", \"content\": user_input}]\n    #   )\n    #   output_text = resp.choices[0].message.content\n\n    latency_ms = round((time.time() - start) * 1000)\n    runs.append({\n        \"example_id\": ex[\"id\"],\n        \"output\": output_text,\n        \"metadata\": {\"model\": \"MODEL_NAME\", \"latency_ms\": latency_ms}\n    })\n    print(f\"  {ex['id']}: {latency_ms}ms\", file=sys.stderr)\n\njson.dump(runs, sys.stdout, indent=2)\n","python",[4846],{"type":53,"tag":74,"props":4847,"children":4848},{"__ignoreMap":391},[4849,4857,4864,4872,4880,4887,4895,4903,4911,4918,4926,4933,4941,4949,4957,4965,4973,4982,4991,5000,5009,5018,5026,5035,5044,5053,5062,5070,5078,5087,5095,5104,5113,5122,5131,5139,5148,5156,5164,5173,5182,5191,5199,5208,5217,5226,5235,5244,5252,5260,5268,5276,5285,5294,5303,5312,5321,5330,5339,5347],{"type":53,"tag":397,"props":4850,"children":4851},{"class":399,"line":400},[4852],{"type":53,"tag":397,"props":4853,"children":4854},{},[4855],{"type":59,"value":4856},"import json, sys, time\n",{"type":53,"tag":397,"props":4858,"children":4859},{"class":399,"line":420},[4860],{"type":53,"tag":397,"props":4861,"children":4862},{"emptyLinePlaceholder":1211},[4863],{"type":59,"value":1214},{"type":53,"tag":397,"props":4865,"children":4866},{"class":399,"line":474},[4867],{"type":53,"tag":397,"props":4868,"children":4869},{},[4870],{"type":59,"value":4871},"examples = json.load(sys.stdin)\n",{"type":53,"tag":397,"props":4873,"children":4874},{"class":399,"line":500},[4875],{"type":53,"tag":397,"props":4876,"children":4877},{},[4878],{"type":59,"value":4879},"runs = []\n",{"type":53,"tag":397,"props":4881,"children":4882},{"class":399,"line":27},[4883],{"type":53,"tag":397,"props":4884,"children":4885},{"emptyLinePlaceholder":1211},[4886],{"type":59,"value":1214},{"type":53,"tag":397,"props":4888,"children":4889},{"class":399,"line":1257},[4890],{"type":53,"tag":397,"props":4891,"children":4892},{},[4893],{"type":59,"value":4894},"for ex in examples:\n",{"type":53,"tag":397,"props":4896,"children":4897},{"class":399,"line":1303},[4898],{"type":53,"tag":397,"props":4899,"children":4900},{},[4901],{"type":59,"value":4902},"    # Inspect the exported JSON to find the right field (e.g. \"input\", \"question\", \"prompt\")\n",{"type":53,"tag":397,"props":4904,"children":4905},{"class":399,"line":1344},[4906],{"type":53,"tag":397,"props":4907,"children":4908},{},[4909],{"type":59,"value":4910},"    user_input = ex.get(\"input\") or ex.get(\"question\") or ex.get(\"prompt\") or str(ex)\n",{"type":53,"tag":397,"props":4912,"children":4913},{"class":399,"line":2007},[4914],{"type":53,"tag":397,"props":4915,"children":4916},{"emptyLinePlaceholder":1211},[4917],{"type":59,"value":1214},{"type":53,"tag":397,"props":4919,"children":4920},{"class":399,"line":2016},[4921],{"type":53,"tag":397,"props":4922,"children":4923},{},[4924],{"type":59,"value":4925},"    start = time.time()\n",{"type":53,"tag":397,"props":4927,"children":4928},{"class":399,"line":2101},[4929],{"type":53,"tag":397,"props":4930,"children":4931},{"emptyLinePlaceholder":1211},[4932],{"type":59,"value":1214},{"type":53,"tag":397,"props":4934,"children":4935},{"class":399,"line":2110},[4936],{"type":53,"tag":397,"props":4937,"children":4938},{},[4939],{"type":59,"value":4940},"    # === CALL THE REAL MODEL API HERE — never fabricate or simulate ===\n",{"type":53,"tag":397,"props":4942,"children":4943},{"class":399,"line":4390},[4944],{"type":53,"tag":397,"props":4945,"children":4946},{},[4947],{"type":59,"value":4948},"    # Uncomment and adapt the provider block the user requested:\n",{"type":53,"tag":397,"props":4950,"children":4951},{"class":399,"line":4420},[4952],{"type":53,"tag":397,"props":4953,"children":4954},{},[4955],{"type":59,"value":4956},"    #\n",{"type":53,"tag":397,"props":4958,"children":4959},{"class":399,"line":4445},[4960],{"type":53,"tag":397,"props":4961,"children":4962},{},[4963],{"type":59,"value":4964},"    # OpenAI (pip install openai  — uses OPENAI_API_KEY env var):\n",{"type":53,"tag":397,"props":4966,"children":4967},{"class":399,"line":4453},[4968],{"type":53,"tag":397,"props":4969,"children":4970},{},[4971],{"type":59,"value":4972},"    #   from openai import OpenAI\n",{"type":53,"tag":397,"props":4974,"children":4976},{"class":399,"line":4975},17,[4977],{"type":53,"tag":397,"props":4978,"children":4979},{},[4980],{"type":59,"value":4981},"    #   resp = OpenAI().chat.completions.create(\n",{"type":53,"tag":397,"props":4983,"children":4985},{"class":399,"line":4984},18,[4986],{"type":53,"tag":397,"props":4987,"children":4988},{},[4989],{"type":59,"value":4990},"    #       model=\"gpt-4o\",\n",{"type":53,"tag":397,"props":4992,"children":4994},{"class":399,"line":4993},19,[4995],{"type":53,"tag":397,"props":4996,"children":4997},{},[4998],{"type":59,"value":4999},"    #       messages=[{\"role\": \"user\", \"content\": user_input}]\n",{"type":53,"tag":397,"props":5001,"children":5003},{"class":399,"line":5002},20,[5004],{"type":53,"tag":397,"props":5005,"children":5006},{},[5007],{"type":59,"value":5008},"    #   )\n",{"type":53,"tag":397,"props":5010,"children":5012},{"class":399,"line":5011},21,[5013],{"type":53,"tag":397,"props":5014,"children":5015},{},[5016],{"type":59,"value":5017},"    #   output_text = resp.choices[0].message.content\n",{"type":53,"tag":397,"props":5019,"children":5021},{"class":399,"line":5020},22,[5022],{"type":53,"tag":397,"props":5023,"children":5024},{},[5025],{"type":59,"value":4956},{"type":53,"tag":397,"props":5027,"children":5029},{"class":399,"line":5028},23,[5030],{"type":53,"tag":397,"props":5031,"children":5032},{},[5033],{"type":59,"value":5034},"    # Anthropic (pip install anthropic  — uses ANTHROPIC_API_KEY env var):\n",{"type":53,"tag":397,"props":5036,"children":5038},{"class":399,"line":5037},24,[5039],{"type":53,"tag":397,"props":5040,"children":5041},{},[5042],{"type":59,"value":5043},"    #   import anthropic\n",{"type":53,"tag":397,"props":5045,"children":5047},{"class":399,"line":5046},25,[5048],{"type":53,"tag":397,"props":5049,"children":5050},{},[5051],{"type":59,"value":5052},"    #   resp = anthropic.Anthropic().messages.create(\n",{"type":53,"tag":397,"props":5054,"children":5056},{"class":399,"line":5055},26,[5057],{"type":53,"tag":397,"props":5058,"children":5059},{},[5060],{"type":59,"value":5061},"    #       model=\"claude-sonnet-4-6\", max_tokens=1024,\n",{"type":53,"tag":397,"props":5063,"children":5065},{"class":399,"line":5064},27,[5066],{"type":53,"tag":397,"props":5067,"children":5068},{},[5069],{"type":59,"value":4999},{"type":53,"tag":397,"props":5071,"children":5073},{"class":399,"line":5072},28,[5074],{"type":53,"tag":397,"props":5075,"children":5076},{},[5077],{"type":59,"value":5008},{"type":53,"tag":397,"props":5079,"children":5081},{"class":399,"line":5080},29,[5082],{"type":53,"tag":397,"props":5083,"children":5084},{},[5085],{"type":59,"value":5086},"    #   output_text = resp.content[0].text\n",{"type":53,"tag":397,"props":5088,"children":5090},{"class":399,"line":5089},30,[5091],{"type":53,"tag":397,"props":5092,"children":5093},{},[5094],{"type":59,"value":4956},{"type":53,"tag":397,"props":5096,"children":5098},{"class":399,"line":5097},31,[5099],{"type":53,"tag":397,"props":5100,"children":5101},{},[5102],{"type":59,"value":5103},"    # Google Gemini (pip install google-genai  — uses GOOGLE_API_KEY env var):\n",{"type":53,"tag":397,"props":5105,"children":5107},{"class":399,"line":5106},32,[5108],{"type":53,"tag":397,"props":5109,"children":5110},{},[5111],{"type":59,"value":5112},"    #   from google import genai\n",{"type":53,"tag":397,"props":5114,"children":5116},{"class":399,"line":5115},33,[5117],{"type":53,"tag":397,"props":5118,"children":5119},{},[5120],{"type":59,"value":5121},"    #   resp = genai.Client().models.generate_content(\n",{"type":53,"tag":397,"props":5123,"children":5125},{"class":399,"line":5124},34,[5126],{"type":53,"tag":397,"props":5127,"children":5128},{},[5129],{"type":59,"value":5130},"    #       model=\"gemini-2.5-pro\", contents=user_input\n",{"type":53,"tag":397,"props":5132,"children":5134},{"class":399,"line":5133},35,[5135],{"type":53,"tag":397,"props":5136,"children":5137},{},[5138],{"type":59,"value":5008},{"type":53,"tag":397,"props":5140,"children":5142},{"class":399,"line":5141},36,[5143],{"type":53,"tag":397,"props":5144,"children":5145},{},[5146],{"type":59,"value":5147},"    #   output_text = resp.text\n",{"type":53,"tag":397,"props":5149,"children":5151},{"class":399,"line":5150},37,[5152],{"type":53,"tag":397,"props":5153,"children":5154},{},[5155],{"type":59,"value":4956},{"type":53,"tag":397,"props":5157,"children":5158},{"class":399,"line":23},[5159],{"type":53,"tag":397,"props":5160,"children":5161},{},[5162],{"type":59,"value":5163},"    # Custom \u002F OpenAI-compatible proxy (pip install openai — uses CUSTOM_BASE_URL + CUSTOM_API_KEY env vars):\n",{"type":53,"tag":397,"props":5165,"children":5167},{"class":399,"line":5166},39,[5168],{"type":53,"tag":397,"props":5169,"children":5170},{},[5171],{"type":59,"value":5172},"    # Use this for Azure OpenAI, NVIDIA NIM, local Ollama, or any OpenAI-compatible endpoint,\n",{"type":53,"tag":397,"props":5174,"children":5176},{"class":399,"line":5175},40,[5177],{"type":53,"tag":397,"props":5178,"children":5179},{},[5180],{"type":59,"value":5181},"    # including a test integration proxy. Matches the `custom` provider in `ax ai-integrations create`.\n",{"type":53,"tag":397,"props":5183,"children":5185},{"class":399,"line":5184},41,[5186],{"type":53,"tag":397,"props":5187,"children":5188},{},[5189],{"type":59,"value":5190},"    #   import os\n",{"type":53,"tag":397,"props":5192,"children":5194},{"class":399,"line":5193},42,[5195],{"type":53,"tag":397,"props":5196,"children":5197},{},[5198],{"type":59,"value":4972},{"type":53,"tag":397,"props":5200,"children":5202},{"class":399,"line":5201},43,[5203],{"type":53,"tag":397,"props":5204,"children":5205},{},[5206],{"type":59,"value":5207},"    #   resp = OpenAI(\n",{"type":53,"tag":397,"props":5209,"children":5211},{"class":399,"line":5210},44,[5212],{"type":53,"tag":397,"props":5213,"children":5214},{},[5215],{"type":59,"value":5216},"    #       base_url=os.environ[\"CUSTOM_BASE_URL\"],          # e.g. https:\u002F\u002Fmy-proxy.example.com\u002Fv1\n",{"type":53,"tag":397,"props":5218,"children":5220},{"class":399,"line":5219},45,[5221],{"type":53,"tag":397,"props":5222,"children":5223},{},[5224],{"type":59,"value":5225},"    #       api_key=os.environ.get(\"CUSTOM_API_KEY\", \"none\"),\n",{"type":53,"tag":397,"props":5227,"children":5229},{"class":399,"line":5228},46,[5230],{"type":53,"tag":397,"props":5231,"children":5232},{},[5233],{"type":59,"value":5234},"    #   ).chat.completions.create(\n",{"type":53,"tag":397,"props":5236,"children":5238},{"class":399,"line":5237},47,[5239],{"type":53,"tag":397,"props":5240,"children":5241},{},[5242],{"type":59,"value":5243},"    #       model=os.environ.get(\"CUSTOM_MODEL\", \"default\"),\n",{"type":53,"tag":397,"props":5245,"children":5247},{"class":399,"line":5246},48,[5248],{"type":53,"tag":397,"props":5249,"children":5250},{},[5251],{"type":59,"value":4999},{"type":53,"tag":397,"props":5253,"children":5255},{"class":399,"line":5254},49,[5256],{"type":53,"tag":397,"props":5257,"children":5258},{},[5259],{"type":59,"value":5008},{"type":53,"tag":397,"props":5261,"children":5263},{"class":399,"line":5262},50,[5264],{"type":53,"tag":397,"props":5265,"children":5266},{},[5267],{"type":59,"value":5017},{"type":53,"tag":397,"props":5269,"children":5271},{"class":399,"line":5270},51,[5272],{"type":53,"tag":397,"props":5273,"children":5274},{"emptyLinePlaceholder":1211},[5275],{"type":59,"value":1214},{"type":53,"tag":397,"props":5277,"children":5279},{"class":399,"line":5278},52,[5280],{"type":53,"tag":397,"props":5281,"children":5282},{},[5283],{"type":59,"value":5284},"    latency_ms = round((time.time() - start) * 1000)\n",{"type":53,"tag":397,"props":5286,"children":5288},{"class":399,"line":5287},53,[5289],{"type":53,"tag":397,"props":5290,"children":5291},{},[5292],{"type":59,"value":5293},"    runs.append({\n",{"type":53,"tag":397,"props":5295,"children":5297},{"class":399,"line":5296},54,[5298],{"type":53,"tag":397,"props":5299,"children":5300},{},[5301],{"type":59,"value":5302},"        \"example_id\": ex[\"id\"],\n",{"type":53,"tag":397,"props":5304,"children":5306},{"class":399,"line":5305},55,[5307],{"type":53,"tag":397,"props":5308,"children":5309},{},[5310],{"type":59,"value":5311},"        \"output\": output_text,\n",{"type":53,"tag":397,"props":5313,"children":5315},{"class":399,"line":5314},56,[5316],{"type":53,"tag":397,"props":5317,"children":5318},{},[5319],{"type":59,"value":5320},"        \"metadata\": {\"model\": \"MODEL_NAME\", \"latency_ms\": latency_ms}\n",{"type":53,"tag":397,"props":5322,"children":5324},{"class":399,"line":5323},57,[5325],{"type":53,"tag":397,"props":5326,"children":5327},{},[5328],{"type":59,"value":5329},"    })\n",{"type":53,"tag":397,"props":5331,"children":5333},{"class":399,"line":5332},58,[5334],{"type":53,"tag":397,"props":5335,"children":5336},{},[5337],{"type":59,"value":5338},"    print(f\"  {ex['id']}: {latency_ms}ms\", file=sys.stderr)\n",{"type":53,"tag":397,"props":5340,"children":5342},{"class":399,"line":5341},59,[5343],{"type":53,"tag":397,"props":5344,"children":5345},{"emptyLinePlaceholder":1211},[5346],{"type":59,"value":1214},{"type":53,"tag":397,"props":5348,"children":5350},{"class":399,"line":5349},60,[5351],{"type":53,"tag":397,"props":5352,"children":5353},{},[5354],{"type":59,"value":5355},"json.dump(runs, sys.stdout, indent=2)\n",{"type":53,"tag":4827,"props":5357,"children":5358},{},[],{"type":53,"tag":70,"props":5360,"children":5361},{},[5362],{"type":59,"value":5363},"Before running:",{"type":59,"value":5365}," install the provider SDK (",{"type":53,"tag":74,"props":5367,"children":5369},{"className":5368},[],[5370],{"type":59,"value":5371},"pip install openai",{"type":59,"value":5373}," \u002F ",{"type":53,"tag":74,"props":5375,"children":5377},{"className":5376},[],[5378],{"type":59,"value":5379},"anthropic",{"type":59,"value":5373},{"type":53,"tag":74,"props":5382,"children":5384},{"className":5383},[],[5385],{"type":59,"value":5386},"google-genai",{"type":59,"value":5388},") and ensure the API key is set as an environment variable in your shell. If you cannot access the API, stop and tell the user what is needed.",{"type":53,"tag":147,"props":5390,"children":5391},{},[5392,5394,5427,5430,5432,5437,5439,5444,5446,5451,5453,5458,5460,5472,5473,5478],{"type":59,"value":5393},"Verify the runs file:",{"type":53,"tag":386,"props":5395,"children":5397},{"className":388,"code":5396,"language":390,"meta":391,"style":391},"python3 -c \"import json; runs=json.load(open('runs.json')); print(f'{len(runs)} runs'); print(json.dumps(runs[0], indent=2))\"\n",[5398],{"type":53,"tag":74,"props":5399,"children":5400},{"__ignoreMap":391},[5401],{"type":53,"tag":397,"props":5402,"children":5403},{"class":399,"line":400},[5404,5409,5414,5418,5423],{"type":53,"tag":397,"props":5405,"children":5406},{"style":404},[5407],{"type":59,"value":5408},"python3",{"type":53,"tag":397,"props":5410,"children":5411},{"style":409},[5412],{"type":59,"value":5413}," -c",{"type":53,"tag":397,"props":5415,"children":5416},{"style":1385},[5417],{"type":59,"value":1721},{"type":53,"tag":397,"props":5419,"children":5420},{"style":409},[5421],{"type":59,"value":5422},"import json; runs=json.load(open('runs.json')); print(f'{len(runs)} runs'); print(json.dumps(runs[0], indent=2))",{"type":53,"tag":397,"props":5424,"children":5425},{"style":1385},[5426],{"type":59,"value":4313},{"type":53,"tag":4827,"props":5428,"children":5429},{},[],{"type":59,"value":5431},"Each run must have ",{"type":53,"tag":74,"props":5433,"children":5435},{"className":5434},[],[5436],{"type":59,"value":1747},{"type":59,"value":5438}," (the dataset row's top-level ",{"type":53,"tag":74,"props":5440,"children":5442},{"className":5441},[],[5443],{"type":59,"value":984},{"type":59,"value":5445},") and ",{"type":53,"tag":74,"props":5447,"children":5449},{"className":5448},[],[5450],{"type":59,"value":1784},{"type":59,"value":5452},". ",{"type":53,"tag":74,"props":5454,"children":5456},{"className":5455},[],[5457],{"type":59,"value":2026},{"type":59,"value":5459}," is optional. ",{"type":53,"tag":70,"props":5461,"children":5462},{},[5463,5465,5470],{"type":59,"value":5464},"Do not put ",{"type":53,"tag":74,"props":5466,"children":5468},{"className":5467},[],[5469],{"type":59,"value":1821},{"type":59,"value":5471}," here",{"type":59,"value":4038},{"type":53,"tag":74,"props":5474,"children":5476},{"className":5475},[],[5477],{"type":59,"value":2806},{"type":59,"value":5479}," ignores them; scores are attached in steps 7–9 below.",{"type":53,"tag":147,"props":5481,"children":5482},{},[5483,5485],{"type":59,"value":5484},"Create the experiment:",{"type":53,"tag":386,"props":5486,"children":5488},{"className":388,"code":5487,"language":390,"meta":391,"style":391},"ax experiments create --name \"gpt-4o-baseline\" --dataset DATASET_NAME --space SPACE --file runs.json\n",[5489],{"type":53,"tag":74,"props":5490,"children":5491},{"__ignoreMap":391},[5492],{"type":53,"tag":397,"props":5493,"children":5494},{"class":399,"line":400},[5495,5499,5503,5507,5511,5515,5519,5523,5527,5531,5535,5539,5543],{"type":53,"tag":397,"props":5496,"children":5497},{"style":404},[5498],{"type":59,"value":224},{"type":53,"tag":397,"props":5500,"children":5501},{"style":409},[5502],{"type":59,"value":412},{"type":53,"tag":397,"props":5504,"children":5505},{"style":409},[5506],{"type":59,"value":2156},{"type":53,"tag":397,"props":5508,"children":5509},{"style":409},[5510],{"type":59,"value":2161},{"type":53,"tag":397,"props":5512,"children":5513},{"style":1385},[5514],{"type":59,"value":1721},{"type":53,"tag":397,"props":5516,"children":5517},{"style":409},[5518],{"type":59,"value":2170},{"type":53,"tag":397,"props":5520,"children":5521},{"style":1385},[5522],{"type":59,"value":1711},{"type":53,"tag":397,"props":5524,"children":5525},{"style":409},[5526],{"type":59,"value":439},{"type":53,"tag":397,"props":5528,"children":5529},{"style":409},[5530],{"type":59,"value":444},{"type":53,"tag":397,"props":5532,"children":5533},{"style":409},[5534],{"type":59,"value":449},{"type":53,"tag":397,"props":5536,"children":5537},{"style":409},[5538],{"type":59,"value":454},{"type":53,"tag":397,"props":5540,"children":5541},{"style":409},[5542],{"type":59,"value":2195},{"type":53,"tag":397,"props":5544,"children":5545},{"style":409},[5546],{"type":59,"value":2200},{"type":53,"tag":147,"props":5548,"children":5549},{},[5550,5552,5558,5561,5566,5568,5572,5574,5579,5581,5586],{"type":59,"value":5551},"Verify: ",{"type":53,"tag":74,"props":5553,"children":5555},{"className":5554},[],[5556],{"type":59,"value":5557},"ax experiments get \"gpt-4o-baseline\" --dataset DATASET_NAME --space SPACE",{"type":53,"tag":4827,"props":5559,"children":5560},{},[],{"type":53,"tag":70,"props":5562,"children":5563},{},[5564],{"type":59,"value":5565},"Attach evaluation scores (required for scores to show in the UI).",{"type":59,"value":5567}," Evaluations do ",{"type":53,"tag":70,"props":5569,"children":5570},{},[5571],{"type":59,"value":2770},{"type":59,"value":5573}," come from the create file — you attach them with ",{"type":53,"tag":74,"props":5575,"children":5577},{"className":5576},[],[5578],{"type":59,"value":3863},{"type":59,"value":5580},", which keys on each run's ",{"type":53,"tag":74,"props":5582,"children":5584},{"className":5583},[],[5585],{"type":59,"value":984},{"type":59,"value":5587}," (assigned at create time), so you must export first to learn those IDs.",{"type":53,"tag":147,"props":5589,"children":5590},{},[5591,5593,5598,5600,5605],{"type":59,"value":5592},"Export the experiment to structured data so you can read each run's ",{"type":53,"tag":74,"props":5594,"children":5596},{"className":5595},[],[5597],{"type":59,"value":984},{"type":59,"value":5599}," alongside its ",{"type":53,"tag":74,"props":5601,"children":5603},{"className":5602},[],[5604],{"type":59,"value":1747},{"type":59,"value":5606},". Confirm that the exported run records include both fields.",{"type":53,"tag":147,"props":5608,"children":5609},{},[5610,5612,5617,5619,5624,5626],{"type":59,"value":5611},"Build the annotation file with structured JSON handling, keyed by ",{"type":53,"tag":74,"props":5613,"children":5615},{"className":5614},[],[5616],{"type":59,"value":3278},{"type":59,"value":5618}," (the run ",{"type":53,"tag":74,"props":5620,"children":5622},{"className":5621},[],[5623],{"type":59,"value":984},{"type":59,"value":5625},"). Score\u002Flabel each run via an LLM-as-judge, a code check, or human review; never fabricate scores. Emit this shape:",{"type":53,"tag":386,"props":5627,"children":5629},{"className":1671,"code":5628,"language":1673,"meta":391,"style":391},"[\n  {\n    \"record_id\": \"RUN_ID_FROM_EXPERIMENT_EXPORT\",\n    \"values\": [\n      { \"name\": \"correctness\", \"score\": 1.0, \"label\": \"correct\" }\n    ]\n  }\n]\n",[5630],{"type":53,"tag":74,"props":5631,"children":5632},{"__ignoreMap":391},[5633,5640,5647,5683,5706,5801,5808,5815],{"type":53,"tag":397,"props":5634,"children":5635},{"class":399,"line":400},[5636],{"type":53,"tag":397,"props":5637,"children":5638},{"style":1385},[5639],{"type":59,"value":1685},{"type":53,"tag":397,"props":5641,"children":5642},{"class":399,"line":420},[5643],{"type":53,"tag":397,"props":5644,"children":5645},{"style":1385},[5646],{"type":59,"value":1693},{"type":53,"tag":397,"props":5648,"children":5649},{"class":399,"line":474},[5650,5654,5658,5662,5666,5670,5675,5679],{"type":53,"tag":397,"props":5651,"children":5652},{"style":1385},[5653],{"type":59,"value":1701},{"type":53,"tag":397,"props":5655,"children":5656},{"style":1704},[5657],{"type":59,"value":3278},{"type":53,"tag":397,"props":5659,"children":5660},{"style":1385},[5661],{"type":59,"value":1711},{"type":53,"tag":397,"props":5663,"children":5664},{"style":1385},[5665],{"type":59,"value":1716},{"type":53,"tag":397,"props":5667,"children":5668},{"style":1385},[5669],{"type":59,"value":1721},{"type":53,"tag":397,"props":5671,"children":5672},{"style":409},[5673],{"type":59,"value":5674},"RUN_ID_FROM_EXPERIMENT_EXPORT",{"type":53,"tag":397,"props":5676,"children":5677},{"style":1385},[5678],{"type":59,"value":1711},{"type":53,"tag":397,"props":5680,"children":5681},{"style":1385},[5682],{"type":59,"value":1735},{"type":53,"tag":397,"props":5684,"children":5685},{"class":399,"line":500},[5686,5690,5694,5698,5702],{"type":53,"tag":397,"props":5687,"children":5688},{"style":1385},[5689],{"type":59,"value":1701},{"type":53,"tag":397,"props":5691,"children":5692},{"style":1704},[5693],{"type":59,"value":3314},{"type":53,"tag":397,"props":5695,"children":5696},{"style":1385},[5697],{"type":59,"value":1711},{"type":53,"tag":397,"props":5699,"children":5700},{"style":1385},[5701],{"type":59,"value":1716},{"type":53,"tag":397,"props":5703,"children":5704},{"style":1385},[5705],{"type":59,"value":3327},{"type":53,"tag":397,"props":5707,"children":5708},{"class":399,"line":27},[5709,5713,5717,5721,5725,5729,5733,5737,5741,5745,5749,5753,5757,5761,5765,5769,5773,5777,5781,5785,5789,5793,5797],{"type":53,"tag":397,"props":5710,"children":5711},{"style":1385},[5712],{"type":59,"value":3335},{"type":53,"tag":397,"props":5714,"children":5715},{"style":1385},[5716],{"type":59,"value":1721},{"type":53,"tag":397,"props":5718,"children":5719},{"style":404},[5720],{"type":59,"value":102},{"type":53,"tag":397,"props":5722,"children":5723},{"style":1385},[5724],{"type":59,"value":1711},{"type":53,"tag":397,"props":5726,"children":5727},{"style":1385},[5728],{"type":59,"value":1716},{"type":53,"tag":397,"props":5730,"children":5731},{"style":1385},[5732],{"type":59,"value":1721},{"type":53,"tag":397,"props":5734,"children":5735},{"style":409},[5736],{"type":59,"value":192},{"type":53,"tag":397,"props":5738,"children":5739},{"style":1385},[5740],{"type":59,"value":1711},{"type":53,"tag":397,"props":5742,"children":5743},{"style":1385},[5744],{"type":59,"value":1894},{"type":53,"tag":397,"props":5746,"children":5747},{"style":1385},[5748],{"type":59,"value":1721},{"type":53,"tag":397,"props":5750,"children":5751},{"style":404},[5752],{"type":59,"value":1903},{"type":53,"tag":397,"props":5754,"children":5755},{"style":1385},[5756],{"type":59,"value":1711},{"type":53,"tag":397,"props":5758,"children":5759},{"style":1385},[5760],{"type":59,"value":1716},{"type":53,"tag":397,"props":5762,"children":5763},{"style":462},[5764],{"type":59,"value":1916},{"type":53,"tag":397,"props":5766,"children":5767},{"style":1385},[5768],{"type":59,"value":1894},{"type":53,"tag":397,"props":5770,"children":5771},{"style":1385},[5772],{"type":59,"value":1721},{"type":53,"tag":397,"props":5774,"children":5775},{"style":404},[5776],{"type":59,"value":1868},{"type":53,"tag":397,"props":5778,"children":5779},{"style":1385},[5780],{"type":59,"value":1711},{"type":53,"tag":397,"props":5782,"children":5783},{"style":1385},[5784],{"type":59,"value":1716},{"type":53,"tag":397,"props":5786,"children":5787},{"style":1385},[5788],{"type":59,"value":1721},{"type":53,"tag":397,"props":5790,"children":5791},{"style":409},[5792],{"type":59,"value":1885},{"type":53,"tag":397,"props":5794,"children":5795},{"style":1385},[5796],{"type":59,"value":1711},{"type":53,"tag":397,"props":5798,"children":5799},{"style":1385},[5800],{"type":59,"value":2004},{"type":53,"tag":397,"props":5802,"children":5803},{"class":399,"line":1257},[5804],{"type":53,"tag":397,"props":5805,"children":5806},{"style":1385},[5807],{"type":59,"value":3526},{"type":53,"tag":397,"props":5809,"children":5810},{"class":399,"line":1303},[5811],{"type":53,"tag":397,"props":5812,"children":5813},{"style":1385},[5814],{"type":59,"value":2107},{"type":53,"tag":397,"props":5816,"children":5817},{"class":399,"line":1344},[5818],{"type":53,"tag":397,"props":5819,"children":5820},{"style":1385},[5821],{"type":59,"value":2116},{"type":53,"tag":147,"props":5823,"children":5824},{},[5825,5827,5833],{"type":59,"value":5826},"Attach the scores with ",{"type":53,"tag":74,"props":5828,"children":5830},{"className":5829},[],[5831],{"type":59,"value":5832},"ax experiments annotate-runs ... --file annotations.json",{"type":59,"value":5834},", then export or inspect the experiment to confirm the evaluations are attached.\nThe scores now render in the experiment view in the Arize UI.",{"type":53,"tag":525,"props":5836,"children":5838},{"id":5837},"compare-two-experiments",[5839],{"type":59,"value":5840},"Compare two experiments",{"type":53,"tag":4622,"props":5842,"children":5843},{},[5844,5970,6060,6145,6189],{"type":53,"tag":147,"props":5845,"children":5846},{},[5847,5849],{"type":59,"value":5848},"Export both experiments:\n",{"type":53,"tag":386,"props":5850,"children":5852},{"className":388,"code":5851,"language":390,"meta":391,"style":391},"ax experiments export \"experiment-a\" --dataset DATASET_NAME --space SPACE --stdout > a.json\nax experiments export \"experiment-b\" --dataset DATASET_NAME --space SPACE --stdout > b.json\n",[5853],{"type":53,"tag":74,"props":5854,"children":5855},{"__ignoreMap":391},[5856,5913],{"type":53,"tag":397,"props":5857,"children":5858},{"class":399,"line":400},[5859,5863,5867,5871,5875,5880,5884,5888,5892,5896,5900,5904,5908],{"type":53,"tag":397,"props":5860,"children":5861},{"style":404},[5862],{"type":59,"value":224},{"type":53,"tag":397,"props":5864,"children":5865},{"style":409},[5866],{"type":59,"value":412},{"type":53,"tag":397,"props":5868,"children":5869},{"style":409},[5870],{"type":59,"value":1175},{"type":53,"tag":397,"props":5872,"children":5873},{"style":1385},[5874],{"type":59,"value":1721},{"type":53,"tag":397,"props":5876,"children":5877},{"style":409},[5878],{"type":59,"value":5879},"experiment-a",{"type":53,"tag":397,"props":5881,"children":5882},{"style":1385},[5883],{"type":59,"value":1711},{"type":53,"tag":397,"props":5885,"children":5886},{"style":409},[5887],{"type":59,"value":439},{"type":53,"tag":397,"props":5889,"children":5890},{"style":409},[5891],{"type":59,"value":444},{"type":53,"tag":397,"props":5893,"children":5894},{"style":409},[5895],{"type":59,"value":449},{"type":53,"tag":397,"props":5897,"children":5898},{"style":409},[5899],{"type":59,"value":454},{"type":53,"tag":397,"props":5901,"children":5902},{"style":409},[5903],{"type":59,"value":1382},{"type":53,"tag":397,"props":5905,"children":5906},{"style":1385},[5907],{"type":59,"value":4821},{"type":53,"tag":397,"props":5909,"children":5910},{"style":409},[5911],{"type":59,"value":5912}," a.json\n",{"type":53,"tag":397,"props":5914,"children":5915},{"class":399,"line":420},[5916,5920,5924,5928,5932,5937,5941,5945,5949,5953,5957,5961,5965],{"type":53,"tag":397,"props":5917,"children":5918},{"style":404},[5919],{"type":59,"value":224},{"type":53,"tag":397,"props":5921,"children":5922},{"style":409},[5923],{"type":59,"value":412},{"type":53,"tag":397,"props":5925,"children":5926},{"style":409},[5927],{"type":59,"value":1175},{"type":53,"tag":397,"props":5929,"children":5930},{"style":1385},[5931],{"type":59,"value":1721},{"type":53,"tag":397,"props":5933,"children":5934},{"style":409},[5935],{"type":59,"value":5936},"experiment-b",{"type":53,"tag":397,"props":5938,"children":5939},{"style":1385},[5940],{"type":59,"value":1711},{"type":53,"tag":397,"props":5942,"children":5943},{"style":409},[5944],{"type":59,"value":439},{"type":53,"tag":397,"props":5946,"children":5947},{"style":409},[5948],{"type":59,"value":444},{"type":53,"tag":397,"props":5950,"children":5951},{"style":409},[5952],{"type":59,"value":449},{"type":53,"tag":397,"props":5954,"children":5955},{"style":409},[5956],{"type":59,"value":454},{"type":53,"tag":397,"props":5958,"children":5959},{"style":409},[5960],{"type":59,"value":1382},{"type":53,"tag":397,"props":5962,"children":5963},{"style":1385},[5964],{"type":59,"value":4821},{"type":53,"tag":397,"props":5966,"children":5967},{"style":409},[5968],{"type":59,"value":5969}," b.json\n",{"type":53,"tag":147,"props":5971,"children":5972},{},[5973,5975,5980,5982],{"type":59,"value":5974},"Compare evaluation scores by ",{"type":53,"tag":74,"props":5976,"children":5978},{"className":5977},[],[5979],{"type":59,"value":1747},{"type":59,"value":5981},":\n",{"type":53,"tag":386,"props":5983,"children":5985},{"className":388,"code":5984,"language":390,"meta":391,"style":391},"# Average correctness score for experiment A\njq '[.[] | .evaluations.correctness.score] | add \u002F length' a.json\n\n# Same for experiment B\njq '[.[] | .evaluations.correctness.score] | add \u002F length' b.json\n",[5986],{"type":53,"tag":74,"props":5987,"children":5988},{"__ignoreMap":391},[5989,5997,6022,6029,6037],{"type":53,"tag":397,"props":5990,"children":5991},{"class":399,"line":400},[5992],{"type":53,"tag":397,"props":5993,"children":5994},{"style":468},[5995],{"type":59,"value":5996},"# Average correctness score for experiment A\n",{"type":53,"tag":397,"props":5998,"children":5999},{"class":399,"line":420},[6000,6005,6009,6014,6018],{"type":53,"tag":397,"props":6001,"children":6002},{"style":404},[6003],{"type":59,"value":6004},"jq",{"type":53,"tag":397,"props":6006,"children":6007},{"style":1385},[6008],{"type":59,"value":1398},{"type":53,"tag":397,"props":6010,"children":6011},{"style":409},[6012],{"type":59,"value":6013},"[.[] | .evaluations.correctness.score] | add \u002F length",{"type":53,"tag":397,"props":6015,"children":6016},{"style":1385},[6017],{"type":59,"value":2462},{"type":53,"tag":397,"props":6019,"children":6020},{"style":409},[6021],{"type":59,"value":5912},{"type":53,"tag":397,"props":6023,"children":6024},{"class":399,"line":474},[6025],{"type":53,"tag":397,"props":6026,"children":6027},{"emptyLinePlaceholder":1211},[6028],{"type":59,"value":1214},{"type":53,"tag":397,"props":6030,"children":6031},{"class":399,"line":500},[6032],{"type":53,"tag":397,"props":6033,"children":6034},{"style":468},[6035],{"type":59,"value":6036},"# Same for experiment B\n",{"type":53,"tag":397,"props":6038,"children":6039},{"class":399,"line":27},[6040,6044,6048,6052,6056],{"type":53,"tag":397,"props":6041,"children":6042},{"style":404},[6043],{"type":59,"value":6004},{"type":53,"tag":397,"props":6045,"children":6046},{"style":1385},[6047],{"type":59,"value":1398},{"type":53,"tag":397,"props":6049,"children":6050},{"style":409},[6051],{"type":59,"value":6013},{"type":53,"tag":397,"props":6053,"children":6054},{"style":1385},[6055],{"type":59,"value":2462},{"type":53,"tag":397,"props":6057,"children":6058},{"style":409},[6059],{"type":59,"value":5969},{"type":53,"tag":147,"props":6061,"children":6062},{},[6063,6065],{"type":59,"value":6064},"Find examples where results differ:\n",{"type":53,"tag":386,"props":6066,"children":6068},{"className":388,"code":6067,"language":390,"meta":391,"style":391},"jq -s '.[0] as $a | .[1][] | . as $run |\n  {\n    example_id: $run.example_id,\n    b_score: $run.evaluations.correctness.score,\n    a_score: ($a[] | select(.example_id == $run.example_id) | .evaluations.correctness.score)\n  }' a.json b.json\n",[6069],{"type":53,"tag":74,"props":6070,"children":6071},{"__ignoreMap":391},[6072,6093,6100,6108,6116,6124],{"type":53,"tag":397,"props":6073,"children":6074},{"class":399,"line":400},[6075,6079,6084,6088],{"type":53,"tag":397,"props":6076,"children":6077},{"style":404},[6078],{"type":59,"value":6004},{"type":53,"tag":397,"props":6080,"children":6081},{"style":409},[6082],{"type":59,"value":6083}," -s",{"type":53,"tag":397,"props":6085,"children":6086},{"style":1385},[6087],{"type":59,"value":1398},{"type":53,"tag":397,"props":6089,"children":6090},{"style":409},[6091],{"type":59,"value":6092},".[0] as $a | .[1][] | . as $run |\n",{"type":53,"tag":397,"props":6094,"children":6095},{"class":399,"line":420},[6096],{"type":53,"tag":397,"props":6097,"children":6098},{"style":409},[6099],{"type":59,"value":1693},{"type":53,"tag":397,"props":6101,"children":6102},{"class":399,"line":474},[6103],{"type":53,"tag":397,"props":6104,"children":6105},{"style":409},[6106],{"type":59,"value":6107},"    example_id: $run.example_id,\n",{"type":53,"tag":397,"props":6109,"children":6110},{"class":399,"line":500},[6111],{"type":53,"tag":397,"props":6112,"children":6113},{"style":409},[6114],{"type":59,"value":6115},"    b_score: $run.evaluations.correctness.score,\n",{"type":53,"tag":397,"props":6117,"children":6118},{"class":399,"line":27},[6119],{"type":53,"tag":397,"props":6120,"children":6121},{"style":409},[6122],{"type":59,"value":6123},"    a_score: ($a[] | select(.example_id == $run.example_id) | .evaluations.correctness.score)\n",{"type":53,"tag":397,"props":6125,"children":6126},{"class":399,"line":1257},[6127,6132,6136,6141],{"type":53,"tag":397,"props":6128,"children":6129},{"style":409},[6130],{"type":59,"value":6131},"  }",{"type":53,"tag":397,"props":6133,"children":6134},{"style":1385},[6135],{"type":59,"value":2462},{"type":53,"tag":397,"props":6137,"children":6138},{"style":409},[6139],{"type":59,"value":6140}," a.json",{"type":53,"tag":397,"props":6142,"children":6143},{"style":409},[6144],{"type":59,"value":5969},{"type":53,"tag":147,"props":6146,"children":6147},{},[6148,6150],{"type":59,"value":6149},"Score distribution per evaluator (pass\u002Ffail\u002Fpartial counts):\n",{"type":53,"tag":386,"props":6151,"children":6153},{"className":388,"code":6152,"language":390,"meta":391,"style":391},"# Count by label for experiment A\njq '[.[] | .evaluations.correctness.label] | group_by(.) | map({label: .[0], count: length})' a.json\n",[6154],{"type":53,"tag":74,"props":6155,"children":6156},{"__ignoreMap":391},[6157,6165],{"type":53,"tag":397,"props":6158,"children":6159},{"class":399,"line":400},[6160],{"type":53,"tag":397,"props":6161,"children":6162},{"style":468},[6163],{"type":59,"value":6164},"# Count by label for experiment A\n",{"type":53,"tag":397,"props":6166,"children":6167},{"class":399,"line":420},[6168,6172,6176,6181,6185],{"type":53,"tag":397,"props":6169,"children":6170},{"style":404},[6171],{"type":59,"value":6004},{"type":53,"tag":397,"props":6173,"children":6174},{"style":1385},[6175],{"type":59,"value":1398},{"type":53,"tag":397,"props":6177,"children":6178},{"style":409},[6179],{"type":59,"value":6180},"[.[] | .evaluations.correctness.label] | group_by(.) | map({label: .[0], count: length})",{"type":53,"tag":397,"props":6182,"children":6183},{"style":1385},[6184],{"type":59,"value":2462},{"type":53,"tag":397,"props":6186,"children":6187},{"style":409},[6188],{"type":59,"value":5912},{"type":53,"tag":147,"props":6190,"children":6191},{},[6192,6194],{"type":59,"value":6193},"Find regressions (examples that passed in A but fail in B):\n",{"type":53,"tag":386,"props":6195,"children":6197},{"className":388,"code":6196,"language":390,"meta":391,"style":391},"jq -s '\n  [.[0][] | select(.evaluations.correctness.label == \"correct\")] as $passed_a |\n  [.[1][] | select(.evaluations.correctness.label != \"correct\") |\n    select(.example_id as $id | $passed_a | any(.example_id == $id))\n  ]\n' a.json b.json\n",[6198],{"type":53,"tag":74,"props":6199,"children":6200},{"__ignoreMap":391},[6201,6217,6225,6233,6241,6249],{"type":53,"tag":397,"props":6202,"children":6203},{"class":399,"line":400},[6204,6208,6212],{"type":53,"tag":397,"props":6205,"children":6206},{"style":404},[6207],{"type":59,"value":6004},{"type":53,"tag":397,"props":6209,"children":6210},{"style":409},[6211],{"type":59,"value":6083},{"type":53,"tag":397,"props":6213,"children":6214},{"style":1385},[6215],{"type":59,"value":6216}," '\n",{"type":53,"tag":397,"props":6218,"children":6219},{"class":399,"line":420},[6220],{"type":53,"tag":397,"props":6221,"children":6222},{"style":409},[6223],{"type":59,"value":6224},"  [.[0][] | select(.evaluations.correctness.label == \"correct\")] as $passed_a |\n",{"type":53,"tag":397,"props":6226,"children":6227},{"class":399,"line":474},[6228],{"type":53,"tag":397,"props":6229,"children":6230},{"style":409},[6231],{"type":59,"value":6232},"  [.[1][] | select(.evaluations.correctness.label != \"correct\") |\n",{"type":53,"tag":397,"props":6234,"children":6235},{"class":399,"line":500},[6236],{"type":53,"tag":397,"props":6237,"children":6238},{"style":409},[6239],{"type":59,"value":6240},"    select(.example_id as $id | $passed_a | any(.example_id == $id))\n",{"type":53,"tag":397,"props":6242,"children":6243},{"class":399,"line":27},[6244],{"type":53,"tag":397,"props":6245,"children":6246},{"style":409},[6247],{"type":59,"value":6248},"  ]\n",{"type":53,"tag":397,"props":6250,"children":6251},{"class":399,"line":1257},[6252,6256,6260],{"type":53,"tag":397,"props":6253,"children":6254},{"style":1385},[6255],{"type":59,"value":2462},{"type":53,"tag":397,"props":6257,"children":6258},{"style":409},[6259],{"type":59,"value":6140},{"type":53,"tag":397,"props":6261,"children":6262},{"style":409},[6263],{"type":59,"value":5969},{"type":53,"tag":66,"props":6265,"children":6266},{},[6267,6272,6274,6280],{"type":53,"tag":70,"props":6268,"children":6269},{},[6270],{"type":59,"value":6271},"Statistical significance note:",{"type":59,"value":6273}," Score comparisons are most reliable with ≥ 30 examples per evaluator. With fewer examples, treat the delta as directional only — a 5% difference on n=10 may be noise. Report sample size alongside scores: ",{"type":53,"tag":74,"props":6275,"children":6277},{"className":6276},[],[6278],{"type":59,"value":6279},"jq 'length' a.json",{"type":59,"value":134},{"type":53,"tag":525,"props":6282,"children":6284},{"id":6283},"download-experiment-results-for-analysis",[6285],{"type":59,"value":6286},"Download experiment results for analysis",{"type":53,"tag":4622,"props":6288,"children":6289},{},[6290,6301,6312],{"type":53,"tag":147,"props":6291,"children":6292},{},[6293,6299],{"type":53,"tag":74,"props":6294,"children":6296},{"className":6295},[],[6297],{"type":59,"value":6298},"ax experiments list --dataset DATASET_NAME --space SPACE",{"type":59,"value":6300}," -- find experiments",{"type":53,"tag":147,"props":6302,"children":6303},{},[6304,6310],{"type":53,"tag":74,"props":6305,"children":6307},{"className":6306},[],[6308],{"type":59,"value":6309},"ax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE",{"type":59,"value":6311}," -- download to file",{"type":53,"tag":147,"props":6313,"children":6314},{},[6315,6317],{"type":59,"value":6316},"Parse: ",{"type":53,"tag":74,"props":6318,"children":6320},{"className":6319},[],[6321],{"type":59,"value":6322},"jq '.[] | {example_id, score: .evaluations.correctness.score}' experiment_*\u002Fruns.json",{"type":53,"tag":525,"props":6324,"children":6326},{"id":6325},"pipe-export-to-other-tools",[6327],{"type":59,"value":6328},"Pipe export to other tools",{"type":53,"tag":386,"props":6330,"children":6332},{"className":388,"code":6331,"language":390,"meta":391,"style":391},"# Count runs\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq 'length'\n\n# Extract all outputs\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq '.[].output'\n\n# Get runs with low scores\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq '[.[] | select(.evaluations.correctness.score \u003C 0.5)]'\n\n# Convert to CSV\nax experiments export EXPERIMENT_NAME --dataset DATASET_NAME --space SPACE --stdout | jq -r '.[] | [.example_id, .output, .evaluations.correctness.score] | @csv'\n",[6333],{"type":53,"tag":74,"props":6334,"children":6335},{"__ignoreMap":391},[6336,6344,6403,6410,6418,6478,6485,6493,6553,6560,6568],{"type":53,"tag":397,"props":6337,"children":6338},{"class":399,"line":400},[6339],{"type":53,"tag":397,"props":6340,"children":6341},{"style":468},[6342],{"type":59,"value":6343},"# Count runs\n",{"type":53,"tag":397,"props":6345,"children":6346},{"class":399,"line":420},[6347,6351,6355,6359,6363,6367,6371,6375,6379,6383,6387,6391,6395,6399],{"type":53,"tag":397,"props":6348,"children":6349},{"style":404},[6350],{"type":59,"value":224},{"type":53,"tag":397,"props":6352,"children":6353},{"style":409},[6354],{"type":59,"value":412},{"type":53,"tag":397,"props":6356,"children":6357},{"style":409},[6358],{"type":59,"value":1175},{"type":53,"tag":397,"props":6360,"children":6361},{"style":409},[6362],{"type":59,"value":1180},{"type":53,"tag":397,"props":6364,"children":6365},{"style":409},[6366],{"type":59,"value":439},{"type":53,"tag":397,"props":6368,"children":6369},{"style":409},[6370],{"type":59,"value":444},{"type":53,"tag":397,"props":6372,"children":6373},{"style":409},[6374],{"type":59,"value":449},{"type":53,"tag":397,"props":6376,"children":6377},{"style":409},[6378],{"type":59,"value":454},{"type":53,"tag":397,"props":6380,"children":6381},{"style":409},[6382],{"type":59,"value":1382},{"type":53,"tag":397,"props":6384,"children":6385},{"style":1385},[6386],{"type":59,"value":1388},{"type":53,"tag":397,"props":6388,"children":6389},{"style":404},[6390],{"type":59,"value":1393},{"type":53,"tag":397,"props":6392,"children":6393},{"style":1385},[6394],{"type":59,"value":1398},{"type":53,"tag":397,"props":6396,"children":6397},{"style":409},[6398],{"type":59,"value":4708},{"type":53,"tag":397,"props":6400,"children":6401},{"style":1385},[6402],{"type":59,"value":1408},{"type":53,"tag":397,"props":6404,"children":6405},{"class":399,"line":474},[6406],{"type":53,"tag":397,"props":6407,"children":6408},{"emptyLinePlaceholder":1211},[6409],{"type":59,"value":1214},{"type":53,"tag":397,"props":6411,"children":6412},{"class":399,"line":500},[6413],{"type":53,"tag":397,"props":6414,"children":6415},{"style":468},[6416],{"type":59,"value":6417},"# Extract all outputs\n",{"type":53,"tag":397,"props":6419,"children":6420},{"class":399,"line":27},[6421,6425,6429,6433,6437,6441,6445,6449,6453,6457,6461,6465,6469,6474],{"type":53,"tag":397,"props":6422,"children":6423},{"style":404},[6424],{"type":59,"value":224},{"type":53,"tag":397,"props":6426,"children":6427},{"style":409},[6428],{"type":59,"value":412},{"type":53,"tag":397,"props":6430,"children":6431},{"style":409},[6432],{"type":59,"value":1175},{"type":53,"tag":397,"props":6434,"children":6435},{"style":409},[6436],{"type":59,"value":1180},{"type":53,"tag":397,"props":6438,"children":6439},{"style":409},[6440],{"type":59,"value":439},{"type":53,"tag":397,"props":6442,"children":6443},{"style":409},[6444],{"type":59,"value":444},{"type":53,"tag":397,"props":6446,"children":6447},{"style":409},[6448],{"type":59,"value":449},{"type":53,"tag":397,"props":6450,"children":6451},{"style":409},[6452],{"type":59,"value":454},{"type":53,"tag":397,"props":6454,"children":6455},{"style":409},[6456],{"type":59,"value":1382},{"type":53,"tag":397,"props":6458,"children":6459},{"style":1385},[6460],{"type":59,"value":1388},{"type":53,"tag":397,"props":6462,"children":6463},{"style":404},[6464],{"type":59,"value":1393},{"type":53,"tag":397,"props":6466,"children":6467},{"style":1385},[6468],{"type":59,"value":1398},{"type":53,"tag":397,"props":6470,"children":6471},{"style":409},[6472],{"type":59,"value":6473},".[].output",{"type":53,"tag":397,"props":6475,"children":6476},{"style":1385},[6477],{"type":59,"value":1408},{"type":53,"tag":397,"props":6479,"children":6480},{"class":399,"line":1257},[6481],{"type":53,"tag":397,"props":6482,"children":6483},{"emptyLinePlaceholder":1211},[6484],{"type":59,"value":1214},{"type":53,"tag":397,"props":6486,"children":6487},{"class":399,"line":1303},[6488],{"type":53,"tag":397,"props":6489,"children":6490},{"style":468},[6491],{"type":59,"value":6492},"# Get runs with low scores\n",{"type":53,"tag":397,"props":6494,"children":6495},{"class":399,"line":1344},[6496,6500,6504,6508,6512,6516,6520,6524,6528,6532,6536,6540,6544,6549],{"type":53,"tag":397,"props":6497,"children":6498},{"style":404},[6499],{"type":59,"value":224},{"type":53,"tag":397,"props":6501,"children":6502},{"style":409},[6503],{"type":59,"value":412},{"type":53,"tag":397,"props":6505,"children":6506},{"style":409},[6507],{"type":59,"value":1175},{"type":53,"tag":397,"props":6509,"children":6510},{"style":409},[6511],{"type":59,"value":1180},{"type":53,"tag":397,"props":6513,"children":6514},{"style":409},[6515],{"type":59,"value":439},{"type":53,"tag":397,"props":6517,"children":6518},{"style":409},[6519],{"type":59,"value":444},{"type":53,"tag":397,"props":6521,"children":6522},{"style":409},[6523],{"type":59,"value":449},{"type":53,"tag":397,"props":6525,"children":6526},{"style":409},[6527],{"type":59,"value":454},{"type":53,"tag":397,"props":6529,"children":6530},{"style":409},[6531],{"type":59,"value":1382},{"type":53,"tag":397,"props":6533,"children":6534},{"style":1385},[6535],{"type":59,"value":1388},{"type":53,"tag":397,"props":6537,"children":6538},{"style":404},[6539],{"type":59,"value":1393},{"type":53,"tag":397,"props":6541,"children":6542},{"style":1385},[6543],{"type":59,"value":1398},{"type":53,"tag":397,"props":6545,"children":6546},{"style":409},[6547],{"type":59,"value":6548},"[.[] | select(.evaluations.correctness.score \u003C 0.5)]",{"type":53,"tag":397,"props":6550,"children":6551},{"style":1385},[6552],{"type":59,"value":1408},{"type":53,"tag":397,"props":6554,"children":6555},{"class":399,"line":2007},[6556],{"type":53,"tag":397,"props":6557,"children":6558},{"emptyLinePlaceholder":1211},[6559],{"type":59,"value":1214},{"type":53,"tag":397,"props":6561,"children":6562},{"class":399,"line":2016},[6563],{"type":53,"tag":397,"props":6564,"children":6565},{"style":468},[6566],{"type":59,"value":6567},"# Convert to CSV\n",{"type":53,"tag":397,"props":6569,"children":6570},{"class":399,"line":2101},[6571,6575,6579,6583,6587,6591,6595,6599,6603,6607,6611,6615,6620,6624,6629],{"type":53,"tag":397,"props":6572,"children":6573},{"style":404},[6574],{"type":59,"value":224},{"type":53,"tag":397,"props":6576,"children":6577},{"style":409},[6578],{"type":59,"value":412},{"type":53,"tag":397,"props":6580,"children":6581},{"style":409},[6582],{"type":59,"value":1175},{"type":53,"tag":397,"props":6584,"children":6585},{"style":409},[6586],{"type":59,"value":1180},{"type":53,"tag":397,"props":6588,"children":6589},{"style":409},[6590],{"type":59,"value":439},{"type":53,"tag":397,"props":6592,"children":6593},{"style":409},[6594],{"type":59,"value":444},{"type":53,"tag":397,"props":6596,"children":6597},{"style":409},[6598],{"type":59,"value":449},{"type":53,"tag":397,"props":6600,"children":6601},{"style":409},[6602],{"type":59,"value":454},{"type":53,"tag":397,"props":6604,"children":6605},{"style":409},[6606],{"type":59,"value":1382},{"type":53,"tag":397,"props":6608,"children":6609},{"style":1385},[6610],{"type":59,"value":1388},{"type":53,"tag":397,"props":6612,"children":6613},{"style":404},[6614],{"type":59,"value":1393},{"type":53,"tag":397,"props":6616,"children":6617},{"style":409},[6618],{"type":59,"value":6619}," -r",{"type":53,"tag":397,"props":6621,"children":6622},{"style":1385},[6623],{"type":59,"value":1398},{"type":53,"tag":397,"props":6625,"children":6626},{"style":409},[6627],{"type":59,"value":6628},".[] | [.example_id, .output, .evaluations.correctness.score] | @csv",{"type":53,"tag":397,"props":6630,"children":6631},{"style":1385},[6632],{"type":59,"value":1408},{"type":53,"tag":136,"props":6634,"children":6636},{"id":6635},"related-skills",[6637],{"type":59,"value":6638},"Related Skills",{"type":53,"tag":143,"props":6640,"children":6641},{},[6642,6659,6677,6692,6707],{"type":53,"tag":147,"props":6643,"children":6644},{},[6645,6650,6652,6657],{"type":53,"tag":70,"props":6646,"children":6647},{},[6648],{"type":59,"value":6649},"arize-dataset",{"type":59,"value":6651},": Create or export the dataset this experiment runs against → use ",{"type":53,"tag":74,"props":6653,"children":6655},{"className":6654},[],[6656],{"type":59,"value":6649},{"type":59,"value":6658}," first",{"type":53,"tag":147,"props":6660,"children":6661},{},[6662,6667,6669,6675],{"type":53,"tag":70,"props":6663,"children":6664},{},[6665],{"type":59,"value":6666},"arize-prompts",{"type":59,"value":6668},": Store and version the prompt template in Prompt Hub (",{"type":53,"tag":74,"props":6670,"children":6672},{"className":6671},[],[6673],{"type":59,"value":6674},"ax prompts",{"type":59,"value":6676},") before or after experiments",{"type":53,"tag":147,"props":6678,"children":6679},{},[6680,6685,6687],{"type":53,"tag":70,"props":6681,"children":6682},{},[6683],{"type":59,"value":6684},"arize-prompt-optimization",{"type":59,"value":6686},": Use experiment results to improve prompts → next step is ",{"type":53,"tag":74,"props":6688,"children":6690},{"className":6689},[],[6691],{"type":59,"value":6684},{"type":53,"tag":147,"props":6693,"children":6694},{},[6695,6700,6702],{"type":53,"tag":70,"props":6696,"children":6697},{},[6698],{"type":59,"value":6699},"arize-trace",{"type":59,"value":6701},": Inspect individual span traces for failing experiment runs → use ",{"type":53,"tag":74,"props":6703,"children":6705},{"className":6704},[],[6706],{"type":59,"value":6699},{"type":53,"tag":147,"props":6708,"children":6709},{},[6710,6715,6717],{"type":53,"tag":70,"props":6711,"children":6712},{},[6713],{"type":59,"value":6714},"arize-link",{"type":59,"value":6716},": Generate clickable UI links to traces from experiment runs → use ",{"type":53,"tag":74,"props":6718,"children":6720},{"className":6719},[],[6721],{"type":59,"value":6714},{"type":53,"tag":136,"props":6723,"children":6725},{"id":6724},"troubleshooting",[6726],{"type":59,"value":6727},"Troubleshooting",{"type":53,"tag":532,"props":6729,"children":6730},{},[6731,6747],{"type":53,"tag":536,"props":6732,"children":6733},{},[6734],{"type":53,"tag":540,"props":6735,"children":6736},{},[6737,6742],{"type":53,"tag":544,"props":6738,"children":6739},{},[6740],{"type":59,"value":6741},"Problem",{"type":53,"tag":544,"props":6743,"children":6744},{},[6745],{"type":59,"value":6746},"Solution",{"type":53,"tag":565,"props":6748,"children":6749},{},[6750,6771,6792,6815,6838,6867,6920,6947,6996,7018],{"type":53,"tag":540,"props":6751,"children":6752},{},[6753,6762],{"type":53,"tag":572,"props":6754,"children":6755},{},[6756],{"type":53,"tag":74,"props":6757,"children":6759},{"className":6758},[],[6760],{"type":59,"value":6761},"ax: command not found",{"type":53,"tag":572,"props":6763,"children":6764},{},[6765,6767],{"type":59,"value":6766},"See ",{"type":53,"tag":254,"props":6768,"children":6769},{"href":256},[6770],{"type":59,"value":256},{"type":53,"tag":540,"props":6772,"children":6773},{},[6774,6782],{"type":53,"tag":572,"props":6775,"children":6776},{},[6777],{"type":53,"tag":74,"props":6778,"children":6780},{"className":6779},[],[6781],{"type":59,"value":267},{"type":53,"tag":572,"props":6783,"children":6784},{},[6785,6787,6791],{"type":59,"value":6786},"API key is wrong, expired, or doesn't have access to this space. Fix the profile using ",{"type":53,"tag":254,"props":6788,"children":6789},{"href":280},[6790],{"type":59,"value":280},{"type":59,"value":134},{"type":53,"tag":540,"props":6793,"children":6794},{},[6795,6804],{"type":53,"tag":572,"props":6796,"children":6797},{},[6798],{"type":53,"tag":74,"props":6799,"children":6801},{"className":6800},[],[6802],{"type":59,"value":6803},"No profile found",{"type":53,"tag":572,"props":6805,"children":6806},{},[6807,6809,6813],{"type":59,"value":6808},"No profile is configured. See ",{"type":53,"tag":254,"props":6810,"children":6811},{"href":280},[6812],{"type":59,"value":280},{"type":59,"value":6814}," to create one.",{"type":53,"tag":540,"props":6816,"children":6817},{},[6818,6827],{"type":53,"tag":572,"props":6819,"children":6820},{},[6821],{"type":53,"tag":74,"props":6822,"children":6824},{"className":6823},[],[6825],{"type":59,"value":6826},"Experiment not found",{"type":53,"tag":572,"props":6828,"children":6829},{},[6830,6832],{"type":59,"value":6831},"Verify experiment name with ",{"type":53,"tag":74,"props":6833,"children":6835},{"className":6834},[],[6836],{"type":59,"value":6837},"ax experiments list --space SPACE",{"type":53,"tag":540,"props":6839,"children":6840},{},[6841,6850],{"type":53,"tag":572,"props":6842,"children":6843},{},[6844],{"type":53,"tag":74,"props":6845,"children":6847},{"className":6846},[],[6848],{"type":59,"value":6849},"Invalid runs file",{"type":53,"tag":572,"props":6851,"children":6852},{},[6853,6854,6859,6860,6865],{"type":59,"value":5431},{"type":53,"tag":74,"props":6855,"children":6857},{"className":6856},[],[6858],{"type":59,"value":1747},{"type":59,"value":2815},{"type":53,"tag":74,"props":6861,"children":6863},{"className":6862},[],[6864],{"type":59,"value":1784},{"type":59,"value":6866}," fields",{"type":53,"tag":540,"props":6868,"children":6869},{},[6870,6879],{"type":53,"tag":572,"props":6871,"children":6872},{},[6873],{"type":53,"tag":74,"props":6874,"children":6876},{"className":6875},[],[6877],{"type":59,"value":6878},"example_id mismatch",{"type":53,"tag":572,"props":6880,"children":6881},{},[6882,6887,6889,6898,6899,6904,6906,6911,6913,6918],{"type":53,"tag":74,"props":6883,"children":6885},{"className":6884},[],[6886],{"type":59,"value":1747},{"type":59,"value":6888}," must be the dataset row's ",{"type":53,"tag":70,"props":6890,"children":6891},{},[6892,6893],{"type":59,"value":2682},{"type":53,"tag":74,"props":6894,"children":6896},{"className":6895},[],[6897],{"type":59,"value":984},{"type":59,"value":2689},{"type":53,"tag":74,"props":6900,"children":6902},{"className":6901},[],[6903],{"type":59,"value":2695},{"type":59,"value":6905}," — not a value nested in the example's fields or ",{"type":53,"tag":74,"props":6907,"children":6909},{"className":6908},[],[6910],{"type":59,"value":2778},{"type":59,"value":6912},". Export the dataset and inspect the top-level ",{"type":53,"tag":74,"props":6914,"children":6916},{"className":6915},[],[6917],{"type":59,"value":984},{"type":59,"value":6919}," field.",{"type":53,"tag":540,"props":6921,"children":6922},{},[6923,6928],{"type":53,"tag":572,"props":6924,"children":6925},{},[6926],{"type":59,"value":6927},"Runs created but no scores \u002F evals in the UI",{"type":53,"tag":572,"props":6929,"children":6930},{},[6931,6933,6938,6940,6945],{"type":59,"value":6932},"Evaluations in the create file are silently ignored. Attach them with ",{"type":53,"tag":74,"props":6934,"children":6936},{"className":6935},[],[6937],{"type":59,"value":2847},{"type":59,"value":6939}," (keyed by run ",{"type":53,"tag":74,"props":6941,"children":6943},{"className":6942},[],[6944],{"type":59,"value":984},{"type":59,"value":6946},") after creating the experiment — see the workflow steps 7–9.",{"type":53,"tag":540,"props":6948,"children":6949},{},[6950,6960],{"type":53,"tag":572,"props":6951,"children":6952},{},[6953,6958],{"type":53,"tag":74,"props":6954,"children":6956},{"className":6955},[],[6957],{"type":59,"value":3863},{"type":59,"value":6959}," reports success but nothing changes",{"type":53,"tag":572,"props":6961,"children":6962},{},[6963,6968,6970,6980,6982,6987,6989,6994],{"type":53,"tag":74,"props":6964,"children":6966},{"className":6965},[],[6967],{"type":59,"value":3278},{"type":59,"value":6969}," must be the ",{"type":53,"tag":70,"props":6971,"children":6972},{},[6973,6975],{"type":59,"value":6974},"run ",{"type":53,"tag":74,"props":6976,"children":6978},{"className":6977},[],[6979],{"type":59,"value":984},{"type":59,"value":6981}," (from ",{"type":53,"tag":74,"props":6983,"children":6985},{"className":6984},[],[6986],{"type":59,"value":1131},{"type":59,"value":6988},"), not the ",{"type":53,"tag":74,"props":6990,"children":6992},{"className":6991},[],[6993],{"type":59,"value":1747},{"type":59,"value":6995},". Unmatched record IDs are silently ignored.",{"type":53,"tag":540,"props":6997,"children":6998},{},[6999,7008],{"type":53,"tag":572,"props":7000,"children":7001},{},[7002],{"type":53,"tag":74,"props":7003,"children":7005},{"className":7004},[],[7006],{"type":59,"value":7007},"No runs found",{"type":53,"tag":572,"props":7009,"children":7010},{},[7011,7013],{"type":59,"value":7012},"Export returned empty -- verify experiment has runs via ",{"type":53,"tag":74,"props":7014,"children":7016},{"className":7015},[],[7017],{"type":59,"value":709},{"type":53,"tag":540,"props":7019,"children":7020},{},[7021,7030],{"type":53,"tag":572,"props":7022,"children":7023},{},[7024],{"type":53,"tag":74,"props":7025,"children":7027},{"className":7026},[],[7028],{"type":59,"value":7029},"Dataset not found",{"type":53,"tag":572,"props":7031,"children":7032},{},[7033,7035],{"type":59,"value":7034},"The linked dataset may have been deleted; check with ",{"type":53,"tag":74,"props":7036,"children":7038},{"className":7037},[],[7039],{"type":59,"value":7040},"ax datasets list",{"type":53,"tag":136,"props":7042,"children":7044},{"id":7043},"save-credentials-for-future-use",[7045],{"type":59,"value":7046},"Save Credentials for Future Use",{"type":53,"tag":66,"props":7048,"children":7049},{},[7050,7051,7055],{"type":59,"value":6766},{"type":53,"tag":254,"props":7052,"children":7053},{"href":280},[7054],{"type":59,"value":280},{"type":59,"value":7056}," § Save Credentials for Future Use.",{"type":53,"tag":7058,"props":7059,"children":7060},"style",{},[7061],{"type":59,"value":7062},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":7064,"total":4390},[7065,7081,7100,7114,7136,7149,7159],{"slug":7066,"name":7066,"fn":7067,"description":7068,"org":7069,"tags":7070,"stars":23,"repoUrl":24,"updatedAt":7080},"arize-admin","manage Arize enterprise user access","Manages Arize users, organizations, spaces, projects, roles, role bindings, resource restrictions, and API keys via the ax CLI. Use for enterprise admin workflows: inviting and offboarding users, onboarding new teams, creating custom roles for SAML\u002FSSO mappings, assigning roles to users, restricting project-level access, and managing service keys for multi-tenant architectures. Covers ax users, ax organizations, ax spaces, ax projects, ax roles, ax role-bindings, and ax api-keys.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7071,7074,7077],{"name":7072,"slug":7073,"type":16},"CLI","cli",{"name":7075,"slug":7076,"type":16},"Operations","operations",{"name":7078,"slug":7079,"type":16},"Permissions","permissions","2026-07-22T05:37:21.991338",{"slug":7082,"name":7082,"fn":7083,"description":7084,"org":7085,"tags":7086,"stars":23,"repoUrl":24,"updatedAt":7099},"arize-ai-provider-integration","manage Arize AI provider integrations","Creates, reads, updates, and deletes Arize AI integrations that store LLM provider credentials used by evaluators and other Arize features. Supports any LLM provider (e.g. OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, Vertex AI, Gemini, NVIDIA NIM). Use when the user mentions AI integration, LLM provider credentials, create integration, list integrations, update credentials, delete integration, or connecting an LLM provider to Arize.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7087,7089,7092,7095,7096],{"name":7088,"slug":5379,"type":16},"Anthropic",{"name":7090,"slug":7091,"type":16},"Azure","azure",{"name":7093,"slug":7094,"type":16},"Integrations","integrations",{"name":14,"slug":15,"type":16},{"name":7097,"slug":7098,"type":16},"OpenAI","openai","2026-07-22T05:37:23.90468",{"slug":7101,"name":7101,"fn":7102,"description":7103,"org":7104,"tags":7105,"stars":23,"repoUrl":24,"updatedAt":7113},"arize-annotation","manage Arize annotation workflows","Creates and manages annotation configs (categorical, continuous, freeform label schemas) and annotation queues (human review workflows) on Arize. Applies human annotations to project spans via the Python SDK. Use when the user mentions annotation config, annotation queue, label schema, human feedback, bulk annotate spans, update_annotations, labeling queue, annotate record, or human review.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7106,7109,7110],{"name":7107,"slug":7108,"type":16},"Data Analysis","data-analysis",{"name":14,"slug":15,"type":16},{"name":7111,"slug":7112,"type":16},"Observability","observability","2026-07-22T05:37:19.010776",{"slug":7115,"name":7115,"fn":7116,"description":7117,"org":7118,"tags":7119,"stars":23,"repoUrl":24,"updatedAt":7135},"arize-compliance-audit","audit AI agents for regulatory compliance","INVOKE THIS SKILL when auditing an AI agent or LLM app for regulatory compliance. Covers EU AI Act, GPAI Code of Practice, GDPR, NIST AI RMF, Colorado AI Act, HIPAA, and ISO 42001. Scans the codebase for compliance gaps, cross-references Arize instrumentation for audit trail coverage, and produces an actionable remediation checklist tailored to the selected frameworks.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7120,7123,7126,7129,7132],{"name":7121,"slug":7122,"type":16},"Audit","audit",{"name":7124,"slug":7125,"type":16},"Compliance","compliance",{"name":7127,"slug":7128,"type":16},"GDPR","gdpr",{"name":7130,"slug":7131,"type":16},"Legal","legal",{"name":7133,"slug":7134,"type":16},"Security","security","2026-07-19T05:39:42.632738",{"slug":6649,"name":6649,"fn":7137,"description":7138,"org":7139,"tags":7140,"stars":23,"repoUrl":24,"updatedAt":7148},"manage Arize datasets and examples","Creates, manages, and queries Arize datasets and examples. Covers dataset CRUD, appending examples, exporting data, and file-based dataset creation using the ax CLI. Use when the user needs test data, evaluation examples, or mentions create dataset, list datasets, export dataset, append examples, dataset version, golden dataset, or test set.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7141,7144,7146,7147],{"name":7142,"slug":7143,"type":16},"Data Engineering","data-engineering",{"name":7145,"slug":35,"type":16},"Datasets",{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},"2026-07-22T05:37:20.943718",{"slug":7150,"name":7150,"fn":7151,"description":7152,"org":7153,"tags":7154,"stars":23,"repoUrl":24,"updatedAt":7158},"arize-evaluator","configure and run Arize evaluations","Handles LLM-as-judge and code evaluator workflows on Arize including creating\u002Fupdating evaluators, running evaluations on spans or experiments, managing tasks, trigger-run operations, column mapping, and continuous monitoring. Use when the user mentions create evaluator, LLM judge, code evaluator, hallucination, faithfulness, correctness, relevance, run eval, score spans, score experiment, trigger-run, column mapping, continuous monitoring, or improve evaluator prompt.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7155,7156,7157],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":7111,"slug":7112,"type":16},"2026-07-25T05:32:37.552903",{"slug":4,"name":4,"fn":5,"description":6,"org":7160,"tags":7161,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7162,7163,7164],{"name":21,"slug":22,"type":16},{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"items":7166,"total":5028},[7167,7181,7190,7202,7214,7224,7236,7246,7257,7263,7271,7277],{"slug":7168,"name":7168,"fn":7169,"description":7170,"org":7171,"tags":7172,"stars":7178,"repoUrl":7179,"updatedAt":7180},"annotate-spans","annotate LLM spans and traces","Write effective, consistent annotations on LLM\u002Fagent spans and traces, and coach the user on annotation practice. Load this whenever you are about to record structured feedback with the `batch_span_annotate` tool, or when the user asks how to annotate, label, score, or review spans\u002Ftraces, build a failure taxonomy, or set up human\u002FLLM review. Do NOT load for: pure analysis with no intent to save feedback (use debug-trace), latency or cost statistics, or prompt authoring (use playground).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7173,7174,7175,7176],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":7111,"slug":7112,"type":16},{"name":7177,"slug":38,"type":16},"Tracing",10513,"https:\u002F\u002Fgithub.com\u002FArize-ai\u002Fphoenix","2026-07-12T08:08:14.140984",{"slug":35,"name":35,"fn":7182,"description":7183,"org":7184,"tags":7185,"stars":7178,"repoUrl":7179,"updatedAt":7189},"reason about Phoenix dataset structure","Understand what a Phoenix dataset is and reason well about its examples, outputs, splits, and how it feeds evaluators and experiments. Load this whenever a dataset is in view or the user asks what a dataset is, how splits work, what an output \"means\", or how datasets relate to experiments and evals. This skill governs the judgment; any tool descriptions govern the mechanics.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7186,7187,7188],{"name":7107,"slug":7108,"type":16},{"name":7145,"slug":35,"type":16},{"name":18,"slug":19,"type":16},"2026-07-12T08:08:21.695457",{"slug":7191,"name":7191,"fn":7192,"description":7193,"org":7194,"tags":7195,"stars":7178,"repoUrl":7179,"updatedAt":7201},"debug-trace","diagnose failures using trace investigation","Diagnose failure modes by systematically investigating traces. Trigger when the user explicitly asks for cross-trace diagnosis: \"what's going wrong?\", \"were there errors?\", \"debug this\", \"where is my agent struggling?\". Do NOT trigger on: (1) advice questions (\"what should I do?\"), (2) statistical questions (\"what's the average latency?\"), (3) summarize requests, (4) trace filtering (\"show me traces with errors\"), (5) vague questions (\"is there a problem?\"), (6) unrelated requests.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7196,7199,7200],{"name":7197,"slug":7198,"type":16},"Debugging","debugging",{"name":7111,"slug":7112,"type":16},{"name":7177,"slug":38,"type":16},"2026-07-12T08:08:10.44243",{"slug":7203,"name":7203,"fn":7204,"description":7205,"org":7206,"tags":7207,"stars":7178,"repoUrl":7179,"updatedAt":7213},"evaluators","author and refine Phoenix evaluators","Author or refine a Phoenix evaluator — code or LLM-as-a-judge — that scores a run's output. Trigger when the user wants to create a new evaluator, improve an existing one's logic or rubric, choose labels, or decide what to measure on a dataset or experiment. Do NOT trigger on: (1) manual prompt drafting (use `playground`), (2) running or comparing experiments themselves (use `experiments`), (3) cross-trace failure diagnosis with no evaluator in scope (use `debug-trace`).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7208,7209,7210],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":7211,"slug":7212,"type":16},"Testing","testing","2026-07-31T05:58:09.13624",{"slug":36,"name":36,"fn":7215,"description":7216,"org":7217,"tags":7218,"stars":7178,"repoUrl":7179,"updatedAt":7223},"run and compare dataset-backed experiments","Run, read, and compare dataset-backed experiments to find evidence that a prompt or pipeline is improving. Trigger when the user wants to iterate over a dataset with experiments, compare experiment runs, read experiment quality\u002Flatency\u002Fcost, or decide whether a change actually helped. Running a prompt over a dataset is implicitly an experiment — load this skill when dataset-backed work begins, before authoring evaluators for the experiment and before starting the recorded run, not only when reading results. Do NOT trigger on: (1) manual prompt drafting with no dataset-backed evaluation in scope (use `playground`), (2) authoring or refining an evaluator's logic or rubric (use `evaluators`), (3) cross-trace failure diagnosis with no experiment in scope (use `debug-trace`).\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7219,7220,7221,7222],{"name":7145,"slug":35,"type":16},{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":7211,"slug":7212,"type":16},"2026-07-12T08:08:11.691477",{"slug":7225,"name":7225,"fn":7226,"description":7227,"org":7228,"tags":7229,"stars":7178,"repoUrl":7179,"updatedAt":7235},"phoenix-graphql","query Phoenix API with GraphQL","Write efficient GraphQL queries against the Phoenix API. Load this skill in two cases: (1) before composing any non-trivial GraphQL query yourself for data analysis (via the `phoenix-gql` bash command) — it contains schema entrypoints and patterns that eliminate the need for introspection; (2) when the user asks for help writing GraphQL queries for their own scripts, tools, or integrations against Phoenix — it covers the endpoint, authentication, and client examples.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7230,7231,7232],{"name":21,"slug":22,"type":16},{"name":7107,"slug":7108,"type":16},{"name":7233,"slug":7234,"type":16},"GraphQL","graphql","2026-07-12T08:08:17.163493",{"slug":7237,"name":7237,"fn":7238,"description":7239,"org":7240,"tags":7241,"stars":7178,"repoUrl":7179,"updatedAt":7245},"playground","author and iterate on prompts in Phoenix","Author, edit, or iterate on prompts in the Phoenix prompt playground, including running experiments over a dataset. Load before any playground tool call, including single-shot prompt rewrites.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7242,7243,7244],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":7211,"slug":7212,"type":16},"2026-07-12T08:08:12.920792",{"slug":7247,"name":7247,"fn":7248,"description":7249,"org":7250,"tags":7251,"stars":7178,"repoUrl":7179,"updatedAt":7256},"span-coding","analyze and code Phoenix spans","Open-code Phoenix spans with PXI-owned notes, recover those notes for axial coding, and promote stable categories into structured annotations. Load this when analyzing spans to discover failure patterns before a taxonomy exists.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7252,7253,7254,7255],{"name":7197,"slug":7198,"type":16},{"name":14,"slug":15,"type":16},{"name":7111,"slug":7112,"type":16},{"name":7177,"slug":38,"type":16},"2026-07-12T08:08:19.597239",{"slug":7066,"name":7066,"fn":7067,"description":7068,"org":7258,"tags":7259,"stars":23,"repoUrl":24,"updatedAt":7080},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7260,7261,7262],{"name":7072,"slug":7073,"type":16},{"name":7075,"slug":7076,"type":16},{"name":7078,"slug":7079,"type":16},{"slug":7082,"name":7082,"fn":7083,"description":7084,"org":7264,"tags":7265,"stars":23,"repoUrl":24,"updatedAt":7099},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7266,7267,7268,7269,7270],{"name":7088,"slug":5379,"type":16},{"name":7090,"slug":7091,"type":16},{"name":7093,"slug":7094,"type":16},{"name":14,"slug":15,"type":16},{"name":7097,"slug":7098,"type":16},{"slug":7101,"name":7101,"fn":7102,"description":7103,"org":7272,"tags":7273,"stars":23,"repoUrl":24,"updatedAt":7113},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7274,7275,7276],{"name":7107,"slug":7108,"type":16},{"name":14,"slug":15,"type":16},{"name":7111,"slug":7112,"type":16},{"slug":7115,"name":7115,"fn":7116,"description":7117,"org":7278,"tags":7279,"stars":23,"repoUrl":24,"updatedAt":7135},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[7280,7281,7282,7283,7284],{"name":7121,"slug":7122,"type":16},{"name":7124,"slug":7125,"type":16},{"name":7127,"slug":7128,"type":16},{"name":7130,"slug":7131,"type":16},{"name":7133,"slug":7134,"type":16}]