eval_results/edbeeching/mixtral-8x7b-instruct-v0.1_merged/v0.1/winogrande/results_2024-04-14T12-24-04.643448.json

{
  "config_general": {
    "lighteval_sha": "?",
    "num_fewshot_seeds": 1,
    "override_batch_size": 1,
    "max_samples": null,
    "job_id": "",
    "start_time": 314635.602497386,
    "end_time": 316560.358900538,
    "total_evaluation_time_secondes": "1924.756403152016",
    "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|winogrande|5": {
      "acc": 0.6164167324388319,
      "acc_stderr": 0.013666275889539017
    },
    "all": {
      "acc": 0.6164167324388319,
      "acc_stderr": 0.013666275889539017
    }
  },
  "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,
      "must_remove_duplicate_docs": null
    }
  },
  "summary_tasks": {
    "leaderboard|winogrande|5": {
      "hashes": {
        "hash_examples": "087d5d1a1afd4c7b",
        "hash_full_prompts": "21d1cf75825bfc31",
        "hash_input_tokens": "74087c8eca355022",
        "hash_cont_tokens": "0dbce81de89a82c0"
      },
      "truncated": 0,
      "non_truncated": 1267,
      "padded": 2391,
      "non_padded": 143,
      "effective_few_shots": 5.0,
      "num_truncated_few_shots": 0
    }
  },
  "summary_general": {
    "hashes": {
      "hash_examples": "b9a49975cc41fab7",
      "hash_full_prompts": "9056d6e5dbbb5d94",
      "hash_input_tokens": "87041500b927b70d",
      "hash_cont_tokens": "366469c3d3fbabf2"
    },
    "truncated": 0,
    "non_truncated": 1267,
    "padded": 2391,
    "non_padded": 143,
    "num_truncated_few_shots": 0
  }
}