{"data":{"kind":"file","path":"README.md","version_id":"ufy0yl9fxg1wh5j1cqu0j3as","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1171,"modified_at":"2026-06-24T01:51:47.677000","content_hash":"345447fcd6962c7836e49c7ecfa7d14756179feab56728ee1f3801f91ee9b61a"},"entries":[],"content":"# vf-v1-test\n\n### Overview\n- **Environment ID**: `vf-v1-test`\n- **Short description**: Tiny self-contained v1 taskset (echo a magic word) for testing the push/install roundtrip.\n- **Tags**: test, single-turn, v1\n\n### Datasets\n- **Primary dataset(s)**: None — 5 tasks are defined inline in `EchoTaskset.load_tasks`.\n- **Source links**: n/a\n- **Split sizes**: 5 tasks (no train/eval split)\n\n### Task\n- **Type**: single-turn\n- **Output format expectations (optional)**: plain text containing the magic word\n- **Rubric overview**: single `echoed` reward — 1.0 if the magic word appears in the reply, else 0.0\n\n### Quickstart\nRun an evaluation with default settings:\n\n```bash\nprime eval run vf-v1-test\n```\n\nConfigure model and sampling:\n\n```bash\nprime eval run vf-v1-test -m openai/gpt-4.1-mini -n 5 -r 1\n```\n\n### Environment Arguments\n\n| Arg | Type | Default | Description |\n| --- | ---- | ------- | ----------- |\n| `num_tasks` | int | `5` | Number of inline echo tasks to load (max 5) |\n\n### Metrics\n\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | Main scalar reward (the `echoed` criterion) |\n| `echoed` | 1.0 if the magic word appears in the reply, else 0.0 |\n","encoding":"utf-8","truncated":false,"total_bytes":1171},"status":null}