{ "config_general": { "lighteval_sha": "?", "num_fewshot_seeds": 1, "override_batch_size": 1, "max_samples": null, "job_id": "", "start_time": 367528.077287332, "end_time": 368058.365594367, "total_evaluation_time_secondes": "530.2883070349926", "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|arc:challenge|25": { "acc": 0.6979522184300341, "acc_stderr": 0.013417519144716417, "acc_norm": 0.7081911262798635, "acc_norm_stderr": 0.013284525292403518 } }, "versions": { "lighteval|arc:challenge|25": 0 }, "config_tasks": { "lighteval|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": [ "lighteval", "arc" ], "original_num_docs": 1172, "effective_num_docs": 1172 } }, "summary_tasks": { "lighteval|arc:challenge|25": { "hashes": { "hash_examples": "17b0cae357c0259e", "hash_full_prompts": "66498cebc864f4ca", "hash_input_tokens": "4cf9f4b8458a5547", "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": "848db321e3b0d4cd", "hash_input_tokens": "ca3626cb628ae01d", "hash_cont_tokens": "6857c42b72c1a62a" }, "truncated": 0, "non_truncated": 1172, "padded": 4660, "non_padded": 27, "num_truncated_few_shots": 0 } }