eval_results/HuggingFaceH4/qwen-1.5-1.8b-dpo/v0.14/hellaswag/results_2024-03-25T21-27-10.180479.json

{
  "config_general": {
    "lighteval_sha": "?",
    "num_fewshot_seeds": 1,
    "override_batch_size": 1,
    "max_samples": null,
    "job_id": "",
    "start_time": 3807581.312783788,
    "end_time": 3807979.900328882,
    "total_evaluation_time_secondes": "398.58754509408027",
    "model_name": "HuggingFaceH4/qwen-1.5-1.8b-dpo",
    "model_sha": "5192a8dac3e74ad2a0528cf921a5652ae4926c20",
    "model_dtype": "torch.bfloat16",
    "model_size": "3.79 GB",
    "config": null
  },
  "results": {
    "leaderboard|hellaswag|10": {
      "acc": 0.4900418243377813,
      "acc_stderr": 0.004988791687322864,
      "acc_norm": 0.5878311093407688,
      "acc_norm_stderr": 0.004912192800263313
    }
  },
  "versions": {
    "leaderboard|hellaswag|10": 0
  },
  "config_tasks": {
    "leaderboard|hellaswag": {
      "name": "hellaswag",
      "prompt_function": "hellaswag_harness",
      "hf_repo": "hellaswag",
      "hf_subset": "default",
      "metric": [
        "loglikelihood_acc",
        "loglikelihood_acc_norm"
      ],
      "hf_avail_splits": [
        "train",
        "test",
        "validation"
      ],
      "evaluation_splits": [
        "validation"
      ],
      "few_shots_split": null,
      "few_shots_select": "random_sampling_from_train",
      "generation_size": -1,
      "stop_sequence": [
        "\n"
      ],
      "output_regex": null,
      "frozen": false,
      "suite": [
        "leaderboard"
      ],
      "original_num_docs": 10042,
      "effective_num_docs": 10042,
      "trust_dataset": true
    }
  },
  "summary_tasks": {
    "leaderboard|hellaswag|10": {
      "hashes": {
        "hash_examples": "31985c805c3a737e",
        "hash_full_prompts": "451e22e8c3aaeeb6",
        "hash_input_tokens": "821cee38b4253cb3",
        "hash_cont_tokens": "3a21ed80a4a7585d"
      },
      "truncated": 0,
      "non_truncated": 10042,
      "padded": 39979,
      "non_padded": 189,
      "effective_few_shots": 10.0,
      "num_truncated_few_shots": 0
    }
  },
  "summary_general": {
    "hashes": {
      "hash_examples": "63bc2cf8bae03fbc",
      "hash_full_prompts": "b2452b7405d0326a",
      "hash_input_tokens": "eb87e19e7255e13b",
      "hash_cont_tokens": "b7bfbb64b00158d2"
    },
    "truncated": 0,
    "non_truncated": 10042,
    "padded": 39979,
    "non_padded": 189,
    "num_truncated_few_shots": 0
  }
}