{ "config_general": { "lighteval_sha": "?", "num_fewshot_seeds": 1, "override_batch_size": 1, "max_samples": null, "job_id": "", "start_time": 2268652.557890831, "end_time": 2268780.24381879, "total_evaluation_time_secondes": "127.68592795915902", "model_name": "HuggingFaceH4/mistral-7b-dpo", "model_sha": "f52a38d6c457274ec93971e0aa6f47ca66c702bc", "model_dtype": "torch.bfloat16", "model_size": "13.99 GB", "config": null }, "results": { "leaderboard|arc:challenge|25": { "acc": 0.5622866894197952, "acc_stderr": 0.014497573881108283, "acc_norm": 0.568259385665529, "acc_norm_stderr": 0.014474591427196204 } }, "versions": { "leaderboard|arc:challenge|25": 0 }, "config_tasks": { "leaderboard|arc:challenge": { "name": "arc:challenge", "prompt_function": "arc", "hf_repo": "ai2_arc", "hf_subset": "ARC-Challenge", "metric": [ "loglikelihood_acc", "loglikelihood_acc_norm_nospace" ], "hf_avail_splits": [ "train", "test" ], "evaluation_splits": [ "test" ], "few_shots_split": null, "few_shots_select": "random_sampling_from_train", "generation_size": 1, "stop_sequence": [ "\n" ], "output_regex": null, "frozen": false, "suite": [ "leaderboard", "arc" ], "original_num_docs": 1172, "effective_num_docs": 1172, "trust_dataset": true } }, "summary_tasks": { "leaderboard|arc:challenge|25": { "hashes": { "hash_examples": "17b0cae357c0259e", "hash_full_prompts": "f42cdfcbcc9b6fa4", "hash_input_tokens": "bedd26d7d7ad2546", "hash_cont_tokens": "8b6f018bc80a3417" }, "truncated": 0, "non_truncated": 1172, "padded": 4660, "non_padded": 27, "effective_few_shots": 25.0, "num_truncated_few_shots": 0 } }, "summary_general": { "hashes": { "hash_examples": "aaa6929c6d3771fb", "hash_full_prompts": "099f2d4c459a2c1d", "hash_input_tokens": "7eee0cf412c3eaf0", "hash_cont_tokens": "6857c42b72c1a62a" }, "truncated": 0, "non_truncated": 1172, "padded": 4660, "non_padded": 27, "num_truncated_few_shots": 0 } }