eval/es_hard__step_9537_1711620130.json

{
  "results": {
    "arc_easy": {
      "acc,none": 0.3265993265993266,
      "acc_stderr,none": 0.009623047038267645,
      "acc_norm,none": 0.3274410774410774,
      "acc_norm_stderr,none": 0.009629415859100605,
      "alias": "arc_easy"
    },
    "lambada_openai": {
      "perplexity,none": 861.0431215826027,
      "perplexity_stderr,none": 49.374856514133086,
      "acc,none": 0.15990685037842034,
      "acc_stderr,none": 0.0051063353047618885,
      "alias": "lambada_openai"
    },
    "piqa": {
      "acc,none": 0.5505984766050055,
      "acc_stderr,none": 0.01160593662415608,
      "acc_norm,none": 0.5391730141458106,
      "acc_norm_stderr,none": 0.011629966056957108,
      "alias": "piqa"
    },
    "sciq": {
      "acc,none": 0.514,
      "acc_stderr,none": 0.015813097547730984,
      "acc_norm,none": 0.511,
      "acc_norm_stderr,none": 0.01581547119529269,
      "alias": "sciq"
    },
    "wikitext": {
      "word_perplexity,none": 14608.989515390149,
      "word_perplexity_stderr,none": "N/A",
      "byte_perplexity,none": 6.00903579422426,
      "byte_perplexity_stderr,none": "N/A",
      "bits_per_byte,none": 2.5871335156471185,
      "bits_per_byte_stderr,none": "N/A",
      "alias": "wikitext"
    },
    "winogrande": {
      "acc,none": 0.5027624309392266,
      "acc_stderr,none": 0.014052271211616436,
      "alias": "winogrande"
    }
  },
  "configs": {
    "arc_easy": {
      "task": "arc_easy",
      "group": [
        "ai2_arc"
      ],
      "dataset_path": "ai2_arc",
      "dataset_name": "ARC-Easy",
      "training_split": "train",
      "validation_split": "validation",
      "test_split": "test",
      "doc_to_text": "Question: {{question}}\nAnswer:",
      "doc_to_target": "{{choices.label.index(answerKey)}}",
      "doc_to_choice": "{{choices.text}}",
      "description": "",
      "target_delimiter": " ",
      "fewshot_delimiter": "\n\n",
      "num_fewshot": 0,
      "metric_list": [
        {
          "metric": "acc",
          "aggregation": "mean",
          "higher_is_better": true
        },
        {
          "metric": "acc_norm",
          "aggregation": "mean",
          "higher_is_better": true
        }
      ],
      "output_type": "multiple_choice",
      "repeats": 1,
      "should_decontaminate": true,
      "doc_to_decontamination_query": "Question: {{question}}\nAnswer:",
      "metadata": {
        "version": 1.0
      }
    },
    "lambada_openai": {
      "task": "lambada_openai",
      "group": [
        "lambada"
      ],
      "dataset_path": "EleutherAI/lambada_openai",
      "dataset_name": "default",
      "test_split": "test",
      "doc_to_text": "{{text.split(' ')[:-1]|join(' ')}}",
      "doc_to_target": "{{' '+text.split(' ')[-1]}}",
      "description": "",
      "target_delimiter": " ",
      "fewshot_delimiter": "\n\n",
      "num_fewshot": 0,
      "metric_list": [
        {
          "metric": "perplexity",
          "aggregation": "perplexity",
          "higher_is_better": false
        },
        {
          "metric": "acc",
          "aggregation": "mean",
          "higher_is_better": true
        }
      ],
      "output_type": "loglikelihood",
      "repeats": 1,
      "should_decontaminate": true,
      "doc_to_decontamination_query": "{{text}}",
      "metadata": {
        "version": 1.0
      }
    },
    "piqa": {
      "task": "piqa",
      "dataset_path": "piqa",
      "training_split": "train",
      "validation_split": "validation",
      "doc_to_text": "Question: {{goal}}\nAnswer:",
      "doc_to_target": "label",
      "doc_to_choice": "{{[sol1, sol2]}}",
      "description": "",
      "target_delimiter": " ",
      "fewshot_delimiter": "\n\n",
      "num_fewshot": 0,
      "metric_list": [
        {
          "metric": "acc",
          "aggregation": "mean",
          "higher_is_better": true
        },
        {
          "metric": "acc_norm",
          "aggregation": "mean",
          "higher_is_better": true
        }
      ],
      "output_type": "multiple_choice",
      "repeats": 1,
      "should_decontaminate": true,
      "doc_to_decontamination_query": "goal",
      "metadata": {
        "version": 1.0
      }
    },
    "sciq": {
      "task": "sciq",
      "dataset_path": "sciq",
      "training_split": "train",
      "validation_split": "validation",
      "test_split": "test",
      "doc_to_text": "{{support.lstrip()}}\nQuestion: {{question}}\nAnswer:",
      "doc_to_target": 3,
      "doc_to_choice": "{{[distractor1, distractor2, distractor3, correct_answer]}}",
      "description": "",
      "target_delimiter": " ",
      "fewshot_delimiter": "\n\n",
      "num_fewshot": 0,
      "metric_list": [
        {
          "metric": "acc",
          "aggregation": "mean",
          "higher_is_better": true
        },
        {
          "metric": "acc_norm",
          "aggregation": "mean",
          "higher_is_better": true
        }
      ],
      "output_type": "multiple_choice",
      "repeats": 1,
      "should_decontaminate": true,
      "doc_to_decontamination_query": "{{support}} {{question}}",
      "metadata": {
        "version": 1.0
      }
    },
    "wikitext": {
      "task": "wikitext",
      "dataset_path": "EleutherAI/wikitext_document_level",
      "dataset_name": "wikitext-2-raw-v1",
      "training_split": "train",
      "validation_split": "validation",
      "test_split": "test",
      "doc_to_text": "",
      "doc_to_target": "<function wikitext_detokenizer at 0x7fb2a52737f0>",
      "process_results": "<function process_results at 0x7fb2a4fb7910>",
      "description": "",
      "target_delimiter": " ",
      "fewshot_delimiter": "\n\n",
      "num_fewshot": 0,
      "metric_list": [
        {
          "metric": "word_perplexity"
        },
        {
          "metric": "byte_perplexity"
        },
        {
          "metric": "bits_per_byte"
        }
      ],
      "output_type": "loglikelihood_rolling",
      "repeats": 1,
      "should_decontaminate": true,
      "doc_to_decontamination_query": "{{page}}",
      "metadata": {
        "version": 2.0
      }
    },
    "winogrande": {
      "task": "winogrande",
      "dataset_path": "winogrande",
      "dataset_name": "winogrande_xl",
      "training_split": "train",
      "validation_split": "validation",
      "doc_to_text": "<function doc_to_text at 0x7fb2a4fd3370>",
      "doc_to_target": "<function doc_to_target at 0x7fb2a4df1480>",
      "doc_to_choice": "<function doc_to_choice at 0x7fb2a4df16c0>",
      "description": "",
      "target_delimiter": " ",
      "fewshot_delimiter": "\n\n",
      "num_fewshot": 0,
      "metric_list": [
        {
          "metric": "acc",
          "aggregation": "mean",
          "higher_is_better": true
        }
      ],
      "output_type": "multiple_choice",
      "repeats": 1,
      "should_decontaminate": true,
      "doc_to_decontamination_query": "sentence",
      "metadata": {
        "version": 1.0
      }
    }
  },
  "versions": {
    "arc_easy": 1.0,
    "lambada_openai": 1.0,
    "piqa": 1.0,
    "sciq": 1.0,
    "wikitext": 2.0,
    "winogrande": 1.0
  },
  "n-shot": {
    "arc_easy": 0,
    "lambada_openai": 0,
    "piqa": 0,
    "sciq": 0,
    "wikitext": 0,
    "winogrande": 0
  },
  "config": {
    "model": "hf",
    "model_args": "pretrained=/netscratch/mostendorff/experiments/pythia-data-ablations/data/continued-pythia-410m/es_hard/hf_checkpoints/global_step9537,dtype=float16",
    "batch_size": "auto",
    "batch_sizes": [
      64
    ],
    "device": "cuda:0",
    "use_cache": null,
    "limit": null,
    "bootstrap_iters": 100000,
    "gen_kwargs": null
  },
  "git_hash": null
}