{"data":{"kind":"file","path":"README.md","version_id":"tsyup5emf0zp9uzels0gvs0p","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1389,"modified_at":"2025-09-02T16:22:18.298000","content_hash":"beaaa47f4d52882327ed7de93e4cd24b5a626c5443a699a9fd9f9dccc2f80d73"},"entries":[],"content":"# haiku\n\n### Overview\n\n- **Environment ID**: `haiku`\n- **Short description**: Write a valid haiku given a user prompt.\n- **Tags**: poetry, haiku\n\n### Datasets\n\n- **Primary dataset(s)**: Instruction Poems from [CheckAI](https://huggingface.co/datasets/checkai/instruction-poems)\n- **Source links**: https://huggingface.co/datasets/checkai/instruction-poems\n- **Split sizes**: Uses the `train` split for evaluation.\n\n### Task\n\n- **Type**: single-turn\n- **Parser**: Parse the response into a list of syllable counts for each line.\n- **Rubric overview**: (1) Difference of line count from 3, (2) Difference of first three line syllable counts from 5, 7, 5.\n\n### Quickstart\n\nRun an evaluation with default settings:\n\n```bash\nuv run vf-eval haiku\n```\n\nConfigure model and sampling:\n\n```bash\nuv run vf-eval haiku -m gpt-4.1-mini -n 20 -r 3 -t 1024 -T 0.7\n```\n\nNotes:\n- Use `-a` / `--env-args` to pass environment-specific configuration as a JSON object.\n\n### Environment Arguments\n\nDocument any supported environment arguments and their meaning. Example:\n\n| Arg | Type | Default | Description |\n| --- | ---- | ------- | ----------- |\n| `max_examples` | int | `-1` | Limit on dataset size (use -1 for all) |\n\n### Metrics\n\nSummarize key metrics your rubric emits and how they’re interpreted.\n\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | Main scalar reward (weighted sum of criteria) |\n","encoding":"utf-8","truncated":false,"total_bytes":1389},"status":null}