| task: humaneval | |
| dataset_path: openai/openai_humaneval | |
| unsafe_code: true | |
| output_type: generate_until | |
| test_split: test | |
| doc_to_text: "{{prompt}}" | |
| doc_to_target: "{{test}}\ncheck({{entry_point}})" | |
| metric_list: | |
| - metric: !function utils.pass_at_k | |
| aggregation: mean | |
| higher_is_better: true | |
| k: [1] | |
| generation_kwargs: | |
| until: | |
| - "\nclass" | |
| - "\ndef" | |
| - "\n#" | |
| - "\nif" | |
| - "\nprint" | |
| max_gen_toks: 1024 | |
| do_sample: false | |
| repeats: 1 | |
| num_fewshot: 0 | |
| filter_list: | |
| - name: "create_test" | |
| filter: | |
| - function: "custom" | |
| filter_fn: !function utils.build_predictions | |
| metadata: | |
| version: 1.0 | |