eval_results/HuggingFaceH4/mistral-7b-odpo/v2.5/truthfulqa/results_2024-05-02T19-09-55.715746.json

{
  "config_general": {
    "lighteval_sha": "?",
    "num_fewshot_seeds": 1,
    "override_batch_size": 1,
    "max_samples": null,
    "job_id": "",
    "start_time": 448865.995378186,
    "end_time": 448971.538318061,
    "total_evaluation_time_secondes": "105.54293987498386",
    "model_name": "HuggingFaceH4/mistral-7b-odpo",
    "model_sha": "6d9ec13fb2f597e578638c45de6fd928228e1053",
    "model_dtype": "torch.bfloat16",
    "model_size": "13.99 GB",
    "config": null
  },
  "results": {
    "leaderboard|truthfulqa:mc|0": {
      "truthfulqa_mc1": 0.31334149326805383,
      "truthfulqa_mc1_stderr": 0.016238065069059605,
      "truthfulqa_mc2": 0.4637030726417336,
      "truthfulqa_mc2_stderr": 0.015195852770394987
    }
  },
  "versions": {
    "leaderboard|truthfulqa:mc|0": 0
  },
  "config_tasks": {
    "leaderboard|truthfulqa:mc": {
      "name": "truthfulqa:mc",
      "prompt_function": "truthful_qa_multiple_choice",
      "hf_repo": "truthful_qa",
      "hf_subset": "multiple_choice",
      "metric": [
        "truthfulqa_mc_metrics"
      ],
      "hf_avail_splits": [
        "validation"
      ],
      "evaluation_splits": [
        "validation"
      ],
      "few_shots_split": null,
      "few_shots_select": null,
      "generation_size": -1,
      "stop_sequence": [
        "\n"
      ],
      "output_regex": null,
      "frozen": false,
      "suite": [
        "leaderboard"
      ],
      "original_num_docs": 817,
      "effective_num_docs": 817,
      "trust_dataset": true
    }
  },
  "summary_tasks": {
    "leaderboard|truthfulqa:mc|0": {
      "hashes": {
        "hash_examples": "36a6d90e75d92d4a",
        "hash_full_prompts": "87686035f32c79c9",
        "hash_input_tokens": "e6efb4cba0ced254",
        "hash_cont_tokens": "9ee6fa47d927a815"
      },
      "truncated": 0,
      "non_truncated": 817,
      "padded": 9429,
      "non_padded": 567,
      "effective_few_shots": 0.0,
      "num_truncated_few_shots": 0
    }
  },
  "summary_general": {
    "hashes": {
      "hash_examples": "aed1dfc67e53d0f2",
      "hash_full_prompts": "3fe6a628d5ea8051",
      "hash_input_tokens": "910f5a1de3b8f8fe",
      "hash_cont_tokens": "ce945fa45483fa47"
    },
    "truncated": 0,
    "non_truncated": 817,
    "padded": 9429,
    "non_padded": 567,
    "num_truncated_few_shots": 0
  }
}