{ "config_general": { "lighteval_sha": "?", "num_fewshot_seeds": 1, "override_batch_size": 1, "max_samples": null, "job_id": "", "start_time": 2255998.622942574, "end_time": 2258673.920192091, "total_evaluation_time_secondes": "2675.297249516938", "model_name": "edbeeching/mixtral-8x7b-instruct-v0.1_merged", "model_sha": "6da964ddd4ac2f1edfa95d2a72082d87f1007cbf", "model_dtype": "torch.bfloat16", "model_size": "87.49 GB", "config": null }, "results": { "leaderboard|truthfulqa:mc|0": { "truthfulqa_mc1": 0.3157894736842105, "truthfulqa_mc1_stderr": 0.016272287957916912, "truthfulqa_mc2": 0.47453692743224557, "truthfulqa_mc2_stderr": 0.015222433901348052 }, "all": { "truthfulqa_mc1": 0.3157894736842105, "truthfulqa_mc1_stderr": 0.016272287957916912, "truthfulqa_mc2": 0.47453692743224557, "truthfulqa_mc2_stderr": 0.015222433901348052 } }, "versions": { "leaderboard|truthfulqa:mc|0": 0 }, "config_tasks": { "leaderboard|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": [ "leaderboard" ], "original_num_docs": 817, "effective_num_docs": 817, "trust_dataset": true, "must_remove_duplicate_docs": null } }, "summary_tasks": { "leaderboard|truthfulqa:mc|0": { "hashes": { "hash_examples": "36a6d90e75d92d4a", "hash_full_prompts": "bc19700dcc192702", "hash_input_tokens": "15d7c94ad93d0bcb", "hash_cont_tokens": "5a8d4bb398b1c3c0" }, "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": "046a77ce3ce1d5d3", "hash_input_tokens": "45a12d916792d80d", "hash_cont_tokens": "3d7ab832fa043aff" }, "truncated": 0, "non_truncated": 817, "padded": 9429, "non_padded": 567, "num_truncated_few_shots": 0 } }