{ "config_general": { "lighteval_sha": "?", "num_fewshot_seeds": 1, "override_batch_size": 1, "max_samples": null, "job_id": "", "start_time": 639324.256860883, "end_time": 640005.247326245, "total_evaluation_time_secondes": "680.9904653619742", "model_name": "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO", "model_sha": "ba6df46252dff6f3978ed416fd95523a3aadb554", "model_dtype": "torch.bfloat16", "model_size": "87.49 GB", "config": null }, "results": { "lighteval|truthfulqa:mc|0": { "truthfulqa_mc1": 0.4773561811505508, "truthfulqa_mc1_stderr": 0.01748554225848965, "truthfulqa_mc2": 0.6475791256422496, "truthfulqa_mc2_stderr": 0.015371085546069103 } }, "versions": { "lighteval|truthfulqa:mc|0": 0 }, "config_tasks": { "lighteval|truthfulqa:mc": { "name": "truthfulqa:mc", "prompt_function": "truthful_qa_multiple_choice", "hf_repo": "truthful_qa", "hf_subset": "multiple_choice", "metric": [ "truthfulqa_mc_metrics" ], "hf_avail_splits": [ "validation" ], "evaluation_splits": [ "validation" ], "few_shots_split": null, "few_shots_select": null, "generation_size": -1, "stop_sequence": [ "\n" ], "output_regex": null, "frozen": false, "suite": [ "lighteval" ], "original_num_docs": 817, "effective_num_docs": 817 } }, "summary_tasks": { "lighteval|truthfulqa:mc|0": { "hashes": { "hash_examples": "36a6d90e75d92d4a", "hash_full_prompts": "17e9d0dc9f923ba3", "hash_input_tokens": "dd712ec831bbf9af", "hash_cont_tokens": "9ee6fa47d927a815" }, "truncated": 0, "non_truncated": 817, "padded": 9429, "non_padded": 567, "effective_few_shots": 0.0, "num_truncated_few_shots": 0 } }, "summary_general": { "hashes": { "hash_examples": "aed1dfc67e53d0f2", "hash_full_prompts": "81a2e5a97bc8b7e3", "hash_input_tokens": "8b0d475877c83266", "hash_cont_tokens": "ce945fa45483fa47" }, "truncated": 0, "non_truncated": 817, "padded": 9429, "non_padded": 567, "num_truncated_few_shots": 0 } }