{ "config_general": { "lighteval_sha": "?", "num_fewshot_seeds": 1, "override_batch_size": 1, "max_samples": null, "job_id": "", "start_time": 1606059.83063072, "end_time": 1606115.08090534, "total_evaluation_time_secondes": "55.2502746200189", "model_name": "HuggingFaceH4/qwen-1.5-1.8b-dpo", "model_sha": "f0063664a3f13e370fd7e02a45b190c0a23e36da", "model_dtype": "torch.bfloat16", "model_size": "3.79 GB", "config": null }, "results": { "leaderboard|winogrande|5": { "acc": 0.574585635359116, "acc_stderr": 0.013895257666646375 } }, "versions": { "leaderboard|winogrande|5": 0 }, "config_tasks": { "leaderboard|winogrande": { "name": "winogrande", "prompt_function": "winogrande", "hf_repo": "winogrande", "hf_subset": "winogrande_xl", "metric": [ "loglikelihood_acc" ], "hf_avail_splits": [ "train", "test", "validation" ], "evaluation_splits": [ "validation" ], "few_shots_split": null, "few_shots_select": "random_sampling", "generation_size": -1, "stop_sequence": [ "\n" ], "output_regex": null, "frozen": false, "suite": [ "leaderboard" ], "original_num_docs": 1267, "effective_num_docs": 1267, "trust_dataset": true } }, "summary_tasks": { "leaderboard|winogrande|5": { "hashes": { "hash_examples": "087d5d1a1afd4c7b", "hash_full_prompts": "29e044bcf40d6a6d", "hash_input_tokens": "1e4162d943d8926d", "hash_cont_tokens": "af83e7e371fcf4d4" }, "truncated": 0, "non_truncated": 1267, "padded": 2356, "non_padded": 178, "effective_few_shots": 5.0, "num_truncated_few_shots": 0 } }, "summary_general": { "hashes": { "hash_examples": "b9a49975cc41fab7", "hash_full_prompts": "2f908b2b9b5ec583", "hash_input_tokens": "b9aa9329c9b8f4e4", "hash_cont_tokens": "cf29ee9ce86b26c4" }, "truncated": 0, "non_truncated": 1267, "padded": 2356, "non_padded": 178, "num_truncated_few_shots": 0 } }