Upload 8 files

EleutherAI-pythia-12b-0shot-shelloutput.txt

@@ -0,0 +1,24 @@
+bootstrapping for stddev: perplexity
+hf (pretrained=EleutherAI/pythia-12b,parallelize=True), limit: None, num_fewshot: 0, batch_size: 4
+|     Task     |Version|Filter|    Metric     | Value |   |Stderr|
+|--------------|-------|------|---------------|------:|---|-----:|
+|arc_challenge |Yaml   |none  |acc            | 0.3157|±  |0.0136|
+|              |       |none  |acc_norm       | 0.3515|±  |0.0140|
+|arc_easy      |Yaml   |none  |acc            | 0.7033|±  |0.0094|
+|              |       |none  |acc_norm       | 0.6372|±  |0.0099|
+|boolq         |Yaml   |none  |acc            | 0.6722|±  |0.0082|
+|hellaswag     |Yaml   |none  |acc            | 0.5038|±  |0.0050|
+|              |       |none  |acc_norm       | 0.6728|±  |0.0047|
+|lambada_openai|Yaml   |none  |perplexity     | 3.9283|±  |0.0838|
+|              |       |none  |acc            | 0.7056|±  |0.0063|
+|openbookqa    |Yaml   |none  |acc            | 0.2640|±  |0.0197|
+|              |       |none  |acc_norm       | 0.3800|±  |0.0217|
+|piqa          |Yaml   |none  |acc            | 0.7612|±  |0.0099|
+|              |       |none  |acc_norm       | 0.7699|±  |0.0098|
+|sciq          |Yaml   |none  |acc            | 0.9040|±  |0.0093|
+|              |       |none  |acc_norm       | 0.8500|±  |0.0113|
+|wikitext      |Yaml   |none  |word_perplexity|16.1038|   |      |
+|              |       |none  |byte_perplexity| 1.5811|   |      |
+|              |       |none  |bits_per_byte  | 0.6610|   |      |
+|winogrande    |Yaml   |none  |acc            | 0.6354|±  |0.0135|
+

EleutherAI-pythia-12b-0shot/results.json

@@ -0,0 +1,404 @@
+{
+  "results": {
+    "arc_challenge": {
+      "acc,none": 0.31569965870307165,
+      "acc_stderr,none": 0.013582571095815295,
+      "acc_norm,none": 0.3515358361774744,
+      "acc_norm_stderr,none": 0.013952413699600931
+    },
+    "arc_easy": {
+      "acc,none": 0.7032828282828283,
+      "acc_stderr,none": 0.009373559492986851,
+      "acc_norm,none": 0.6372053872053872,
+      "acc_norm_stderr,none": 0.009865936757013938
+    },
+    "boolq": {
+      "acc,none": 0.67217125382263,
+      "acc_stderr,none": 0.008210243237673385
+    },
+    "hellaswag": {
+      "acc,none": 0.5037841067516431,
+      "acc_stderr,none": 0.004989638507409938,
+      "acc_norm,none": 0.6727743477394941,
+      "acc_norm_stderr,none": 0.004682414968323615
+    },
+    "lambada_openai": {
+      "perplexity,none": 3.928303151284302,
+      "perplexity_stderr,none": 0.08380809965264784,
+      "acc,none": 0.7056083834659421,
+      "acc_stderr,none": 0.006349750457397497
+    },
+    "openbookqa": {
+      "acc,none": 0.264,
+      "acc_stderr,none": 0.019732885585922098,
+      "acc_norm,none": 0.38,
+      "acc_norm_stderr,none": 0.02172888143870172
+    },
+    "piqa": {
+      "acc,none": 0.7611534276387377,
+      "acc_stderr,none": 0.009948120385337496,
+      "acc_norm,none": 0.7698585418933623,
+      "acc_norm_stderr,none": 0.009820832826839813
+    },
+    "sciq": {
+      "acc,none": 0.904,
+      "acc_stderr,none": 0.009320454434783203,
+      "acc_norm,none": 0.85,
+      "acc_norm_stderr,none": 0.01129723982340931
+    },
+    "wikitext": {
+      "word_perplexity,none": 16.10381134238307,
+      "byte_perplexity,none": 1.5811322783219102,
+      "bits_per_byte,none": 0.6609580693383097
+    },
+    "winogrande": {
+      "acc,none": 0.6353591160220995,
+      "acc_stderr,none": 0.013527746622429842
+    }
+  },
+  "configs": {
+    "arc_challenge": {
+      "task": "arc_challenge",
+      "group": [
+        "ai2_arc",
+        "multiple_choice"
+      ],
+      "dataset_path": "ai2_arc",
+      "dataset_name": "ARC-Challenge",
+      "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:"
+    },
+    "arc_easy": {
+      "task": "arc_easy",
+      "group": [
+        "ai2_arc",
+        "multiple_choice"
+      ],
+      "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:"
+    },
+    "boolq": {
+      "task": "boolq",
+      "group": [
+        "super-glue-lm-eval-v1"
+      ],
+      "dataset_path": "super_glue",
+      "dataset_name": "boolq",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "{{passage}}\nQuestion: {{question}}?\nAnswer:",
+      "doc_to_target": "label",
+      "doc_to_choice": [
+        "no",
+        "yes"
+      ],
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 0,
+      "metric_list": [
+        {
+          "metric": "acc"
+        }
+      ],
+      "output_type": "multiple_choice",
+      "repeats": 1,
+      "should_decontaminate": true,
+      "doc_to_decontamination_query": "passage"
+    },
+    "hellaswag": {
+      "task": "hellaswag",
+      "group": [
+        "multiple_choice"
+      ],
+      "dataset_path": "hellaswag",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "{% set text = activity_label ~ ': ' ~ ctx_a ~ ' ' ~ ctx_b.capitalize() %}{{text|trim|replace(' [title]', '. ')|regex_replace('\\[.*?\\]', '')|replace('  ', ' ')}}",
+      "doc_to_target": "{{label}}",
+      "doc_to_choice": "{{endings|map('trim')|map('replace', ' [title]', '. ')|map('regex_replace', '\\[.*?\\]', '')|map('replace', '  ', ' ')|list}}",
+      "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": false
+    },
+    "lambada_openai": {
+      "task": "lambada_openai",
+      "group": [
+        "lambada",
+        "loglikelihood",
+        "perplexity"
+      ],
+      "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}}"
+    },
+    "openbookqa": {
+      "task": "openbookqa",
+      "group": [
+        "multiple_choice"
+      ],
+      "dataset_path": "openbookqa",
+      "dataset_name": "main",
+      "training_split": "train",
+      "validation_split": "validation",
+      "test_split": "test",
+      "doc_to_text": "question_stem",
+      "doc_to_target": "{{choices.label.index(answerKey.lstrip())}}",
+      "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_stem"
+    },
+    "piqa": {
+      "task": "piqa",
+      "group": [
+        "multiple_choice"
+      ],
+      "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"
+    },
+    "sciq": {
+      "task": "sciq",
+      "group": [
+        "multiple_choice"
+      ],
+      "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}}"
+    },
+    "wikitext": {
+      "task": "wikitext",
+      "group": [
+        "perplexity",
+        "loglikelihood_rolling"
+      ],
+      "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 0x7f1b6d8b9120>",
+      "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}}"
+    },
+    "winogrande": {
+      "task": "winogrande",
+      "dataset_path": "winogrande",
+      "dataset_name": "winogrande_xl",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "<function doc_to_text at 0x7f1b6d8b9360>",
+      "doc_to_target": "<function doc_to_target at 0x7f1b6d8b96c0>",
+      "doc_to_choice": "<function doc_to_choice at 0x7f1b6d8b9a20>",
+      "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"
+    }
+  },
+  "versions": {
+    "arc_challenge": "Yaml",
+    "arc_easy": "Yaml",
+    "boolq": "Yaml",
+    "hellaswag": "Yaml",
+    "lambada_openai": "Yaml",
+    "openbookqa": "Yaml",
+    "piqa": "Yaml",
+    "sciq": "Yaml",
+    "wikitext": "Yaml",
+    "winogrande": "Yaml"
+  },
+  "config": {
+    "model": "hf",
+    "model_args": "pretrained=EleutherAI/pythia-12b,parallelize=True",
+    "batch_size": "4",
+    "batch_sizes": [],
+    "device": null,
+    "use_cache": null,
+    "limit": null,
+    "bootstrap_iters": 100000
+  },
+  "git_hash": "d1a44c8"
+}

EleutherAI-pythia-12b-5shot-shelloutput.txt

@@ -0,0 +1,24 @@
+bootstrapping for stddev: perplexity
+hf (pretrained=EleutherAI/pythia-12b,parallelize=True), limit: None, num_fewshot: 5, batch_size: 4
+|     Task     |Version|Filter|    Metric     | Value |   |Stderr|
+|--------------|-------|------|---------------|------:|---|-----:|
+|arc_challenge |Yaml   |none  |acc            | 0.3686|±  |0.0141|
+|              |       |none  |acc_norm       | 0.4010|±  |0.0143|
+|arc_easy      |Yaml   |none  |acc            | 0.7088|±  |0.0093|
+|              |       |none  |acc_norm       | 0.7146|±  |0.0093|
+|boolq         |Yaml   |none  |acc            | 0.6795|±  |0.0082|
+|hellaswag     |Yaml   |none  |acc            | 0.5043|±  |0.0050|
+|              |       |none  |acc_norm       | 0.6830|±  |0.0046|
+|lambada_openai|Yaml   |none  |perplexity     | 4.7835|±  |0.1073|
+|              |       |none  |acc            | 0.6625|±  |0.0066|
+|openbookqa    |Yaml   |none  |acc            | 0.2940|±  |0.0204|
+|              |       |none  |acc_norm       | 0.3980|±  |0.0219|
+|piqa          |Yaml   |none  |acc            | 0.7699|±  |0.0098|
+|              |       |none  |acc_norm       | 0.7731|±  |0.0098|
+|sciq          |Yaml   |none  |acc            | 0.9480|±  |0.0070|
+|              |       |none  |acc_norm       | 0.9510|±  |0.0068|
+|wikitext      |Yaml   |none  |word_perplexity|16.1038|   |      |
+|              |       |none  |byte_perplexity| 1.5811|   |      |
+|              |       |none  |bits_per_byte  | 0.6610|   |      |
+|winogrande    |Yaml   |none  |acc            | 0.6354|±  |0.0135|
+

EleutherAI-pythia-12b-5shot/results.json

@@ -0,0 +1,404 @@
+{
+  "results": {
+    "arc_challenge": {
+      "acc,none": 0.36860068259385664,
+      "acc_stderr,none": 0.014097810678042194,
+      "acc_norm,none": 0.40102389078498296,
+      "acc_norm_stderr,none": 0.014322255790719867
+    },
+    "arc_easy": {
+      "acc,none": 0.7087542087542088,
+      "acc_stderr,none": 0.009322788837938856,
+      "acc_norm,none": 0.7146464646464646,
+      "acc_norm_stderr,none": 0.00926628058499775
+    },
+    "boolq": {
+      "acc,none": 0.6795107033639144,
+      "acc_stderr,none": 0.008162016261049398
+    },
+    "hellaswag": {
+      "acc,none": 0.504282015534754,
+      "acc_stderr,none": 0.0049895984262495335,
+      "acc_norm,none": 0.6830312686715794,
+      "acc_norm_stderr,none": 0.004643441945489853
+    },
+    "lambada_openai": {
+      "perplexity,none": 4.783522630921447,
+      "perplexity_stderr,none": 0.10727737476272488,
+      "acc,none": 0.6625266834853484,
+      "acc_stderr,none": 0.006587694938528712
+    },
+    "openbookqa": {
+      "acc,none": 0.294,
+      "acc_stderr,none": 0.020395095484936614,
+      "acc_norm,none": 0.398,
+      "acc_norm_stderr,none": 0.02191237788577998
+    },
+    "piqa": {
+      "acc,none": 0.7698585418933623,
+      "acc_stderr,none": 0.009820832826839817,
+      "acc_norm,none": 0.7731229597388466,
+      "acc_norm_stderr,none": 0.00977158425921514
+    },
+    "sciq": {
+      "acc,none": 0.948,
+      "acc_stderr,none": 0.0070246242138171325,
+      "acc_norm,none": 0.951,
+      "acc_norm_stderr,none": 0.006829761756140924
+    },
+    "wikitext": {
+      "word_perplexity,none": 16.10381134238307,
+      "byte_perplexity,none": 1.5811322783219102,
+      "bits_per_byte,none": 0.6609580693383097
+    },
+    "winogrande": {
+      "acc,none": 0.6353591160220995,
+      "acc_stderr,none": 0.013527746622429842
+    }
+  },
+  "configs": {
+    "arc_challenge": {
+      "task": "arc_challenge",
+      "group": [
+        "ai2_arc",
+        "multiple_choice"
+      ],
+      "dataset_path": "ai2_arc",
+      "dataset_name": "ARC-Challenge",
+      "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": 5,
+      "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:"
+    },
+    "arc_easy": {
+      "task": "arc_easy",
+      "group": [
+        "ai2_arc",
+        "multiple_choice"
+      ],
+      "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": 5,
+      "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:"
+    },
+    "boolq": {
+      "task": "boolq",
+      "group": [
+        "super-glue-lm-eval-v1"
+      ],
+      "dataset_path": "super_glue",
+      "dataset_name": "boolq",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "{{passage}}\nQuestion: {{question}}?\nAnswer:",
+      "doc_to_target": "label",
+      "doc_to_choice": [
+        "no",
+        "yes"
+      ],
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "metric_list": [
+        {
+          "metric": "acc"
+        }
+      ],
+      "output_type": "multiple_choice",
+      "repeats": 1,
+      "should_decontaminate": true,
+      "doc_to_decontamination_query": "passage"
+    },
+    "hellaswag": {
+      "task": "hellaswag",
+      "group": [
+        "multiple_choice"
+      ],
+      "dataset_path": "hellaswag",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "{% set text = activity_label ~ ': ' ~ ctx_a ~ ' ' ~ ctx_b.capitalize() %}{{text|trim|replace(' [title]', '. ')|regex_replace('\\[.*?\\]', '')|replace('  ', ' ')}}",
+      "doc_to_target": "{{label}}",
+      "doc_to_choice": "{{endings|map('trim')|map('replace', ' [title]', '. ')|map('regex_replace', '\\[.*?\\]', '')|map('replace', '  ', ' ')|list}}",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "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": false
+    },
+    "lambada_openai": {
+      "task": "lambada_openai",
+      "group": [
+        "lambada",
+        "loglikelihood",
+        "perplexity"
+      ],
+      "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": 5,
+      "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}}"
+    },
+    "openbookqa": {
+      "task": "openbookqa",
+      "group": [
+        "multiple_choice"
+      ],
+      "dataset_path": "openbookqa",
+      "dataset_name": "main",
+      "training_split": "train",
+      "validation_split": "validation",
+      "test_split": "test",
+      "doc_to_text": "question_stem",
+      "doc_to_target": "{{choices.label.index(answerKey.lstrip())}}",
+      "doc_to_choice": "{{choices.text}}",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "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_stem"
+    },
+    "piqa": {
+      "task": "piqa",
+      "group": [
+        "multiple_choice"
+      ],
+      "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": 5,
+      "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"
+    },
+    "sciq": {
+      "task": "sciq",
+      "group": [
+        "multiple_choice"
+      ],
+      "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": 5,
+      "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}}"
+    },
+    "wikitext": {
+      "task": "wikitext",
+      "group": [
+        "perplexity",
+        "loglikelihood_rolling"
+      ],
+      "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 0x7f9e93ad1120>",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "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}}"
+    },
+    "winogrande": {
+      "task": "winogrande",
+      "dataset_path": "winogrande",
+      "dataset_name": "winogrande_xl",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "<function doc_to_text at 0x7f9e93ad1360>",
+      "doc_to_target": "<function doc_to_target at 0x7f9e93ad16c0>",
+      "doc_to_choice": "<function doc_to_choice at 0x7f9e93ad1a20>",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "metric_list": [
+        {
+          "metric": "acc",
+          "aggregation": "mean",
+          "higher_is_better": true
+        }
+      ],
+      "output_type": "multiple_choice",
+      "repeats": 1,
+      "should_decontaminate": true,
+      "doc_to_decontamination_query": "sentence"
+    }
+  },
+  "versions": {
+    "arc_challenge": "Yaml",
+    "arc_easy": "Yaml",
+    "boolq": "Yaml",
+    "hellaswag": "Yaml",
+    "lambada_openai": "Yaml",
+    "openbookqa": "Yaml",
+    "piqa": "Yaml",
+    "sciq": "Yaml",
+    "wikitext": "Yaml",
+    "winogrande": "Yaml"
+  },
+  "config": {
+    "model": "hf",
+    "model_args": "pretrained=EleutherAI/pythia-12b,parallelize=True",
+    "batch_size": "4",
+    "batch_sizes": [],
+    "device": null,
+    "use_cache": null,
+    "limit": null,
+    "bootstrap_iters": 100000
+  },
+  "git_hash": "d1a44c8"
+}

sft-pythia-12b-0shot-shelloutput.txt

@@ -0,0 +1,24 @@
+bootstrapping for stddev: perplexity
+hf (pretrained=lomahony/eleuther-pythia12b-hh-sft,parallelize=True), limit: None, num_fewshot: 0, batch_size: 16
+|     Task     |Version|Filter|    Metric     | Value |   |Stderr|
+|--------------|-------|------|---------------|------:|---|-----:|
+|arc_challenge |Yaml   |none  |acc            | 0.3106|±  |0.0135|
+|              |       |none  |acc_norm       | 0.3464|±  |0.0139|
+|arc_easy      |Yaml   |none  |acc            | 0.7012|±  |0.0094|
+|              |       |none  |acc_norm       | 0.6187|±  |0.0100|
+|boolq         |Yaml   |none  |acc            | 0.6954|±  |0.0080|
+|hellaswag     |Yaml   |none  |acc            | 0.5056|±  |0.0050|
+|              |       |none  |acc_norm       | 0.6668|±  |0.0047|
+|lambada_openai|Yaml   |none  |perplexity     | 3.5325|±  |0.0776|
+|              |       |none  |acc            | 0.7105|±  |0.0063|
+|openbookqa    |Yaml   |none  |acc            | 0.2760|±  |0.0200|
+|              |       |none  |acc_norm       | 0.3820|±  |0.0218|
+|piqa          |Yaml   |none  |acc            | 0.7633|±  |0.0099|
+|              |       |none  |acc_norm       | 0.7644|±  |0.0099|
+|sciq          |Yaml   |none  |acc            | 0.9060|±  |0.0092|
+|              |       |none  |acc_norm       | 0.8440|±  |0.0115|
+|wikitext      |Yaml   |none  |word_perplexity|16.5611|   |      |
+|              |       |none  |byte_perplexity| 1.5884|   |      |
+|              |       |none  |bits_per_byte  | 0.6676|   |      |
+|winogrande    |Yaml   |none  |acc            | 0.6346|±  |0.0135|
+

sft-pythia-12b-0shot/results.json

@@ -0,0 +1,388 @@
+{
+  "results": {
+    "arc_challenge": {
+      "acc,none": 0.310580204778157,
+      "acc_stderr,none": 0.013522292098053069,
+      "acc_norm,none": 0.3464163822525597,
+      "acc_norm_stderr,none": 0.013905011180063242
+    },
+    "arc_easy": {
+      "acc,none": 0.7011784511784511,
+      "acc_stderr,none": 0.009392656275408732,
+      "acc_norm,none": 0.6186868686868687,
+      "acc_norm_stderr,none": 0.00996654249717102
+    },
+    "boolq": {
+      "acc,none": 0.6954128440366972,
+      "acc_stderr,none": 0.00804951448892039
+    },
+    "hellaswag": {
+      "acc,none": 0.5055765783708425,
+      "acc_stderr,none": 0.004989471055090951,
+      "acc_norm,none": 0.6667994423421629,
+      "acc_norm_stderr,none": 0.004703942346762236
+    },
+    "lambada_openai": {
+      "perplexity,none": 3.5324751127269423,
+      "perplexity_stderr,none": 0.07759679763422017,
+      "acc,none": 0.7104599262565496,
+      "acc_stderr,none": 0.006318823234213228
+    },
+    "openbookqa": {
+      "acc,none": 0.276,
+      "acc_stderr,none": 0.02001121929807353,
+      "acc_norm,none": 0.382,
+      "acc_norm_stderr,none": 0.021750820591250827
+    },
+    "piqa": {
+      "acc,none": 0.7633297062023939,
+      "acc_stderr,none": 0.009916841655042807,
+      "acc_norm,none": 0.764417845484222,
+      "acc_norm_stderr,none": 0.009901067586473916
+    },
+    "sciq": {
+      "acc,none": 0.906,
+      "acc_stderr,none": 0.009233052000787728,
+      "acc_norm,none": 0.844,
+      "acc_norm_stderr,none": 0.011480235006122356
+    },
+    "wikitext": {
+      "word_perplexity,none": 16.56112447245529,
+      "byte_perplexity,none": 1.588448103407034,
+      "bits_per_byte,none": 0.6676179561638627
+    },
+    "winogrande": {
+      "acc,none": 0.6345698500394633,
+      "acc_stderr,none": 0.013533965097638798
+    }
+  },
+  "configs": {
+    "arc_challenge": {
+      "task": "arc_challenge",
+      "group": [
+        "ai2_arc"
+      ],
+      "dataset_path": "ai2_arc",
+      "dataset_name": "ARC-Challenge",
+      "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:"
+    },
+    "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:"
+    },
+    "boolq": {
+      "task": "boolq",
+      "group": [
+        "super-glue-lm-eval-v1"
+      ],
+      "dataset_path": "super_glue",
+      "dataset_name": "boolq",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "{{passage}}\nQuestion: {{question}}?\nAnswer:",
+      "doc_to_target": "label",
+      "doc_to_choice": [
+        "no",
+        "yes"
+      ],
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 0,
+      "metric_list": [
+        {
+          "metric": "acc"
+        }
+      ],
+      "output_type": "multiple_choice",
+      "repeats": 1,
+      "should_decontaminate": true,
+      "doc_to_decontamination_query": "passage"
+    },
+    "hellaswag": {
+      "task": "hellaswag",
+      "group": [
+        "multiple_choice"
+      ],
+      "dataset_path": "hellaswag",
+      "training_split": "train",
+      "validation_split": "validation",
+      "process_docs": "<function process_docs at 0x7fa0005d55e0>",
+      "doc_to_text": "{{query}}",
+      "doc_to_target": "{{label}}",
+      "doc_to_choice": "{{choices}}",
+      "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": false
+    },
+    "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}}"
+    },
+    "openbookqa": {
+      "task": "openbookqa",
+      "dataset_path": "openbookqa",
+      "dataset_name": "main",
+      "training_split": "train",
+      "validation_split": "validation",
+      "test_split": "test",
+      "doc_to_text": "question_stem",
+      "doc_to_target": "{{choices.label.index(answerKey.lstrip())}}",
+      "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_stem"
+    },
+    "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"
+    },
+    "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}}"
+    },
+    "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 0x7f9fff5441f0>",
+      "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}}"
+    },
+    "winogrande": {
+      "task": "winogrande",
+      "dataset_path": "winogrande",
+      "dataset_name": "winogrande_xl",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "<function doc_to_text at 0x7f9fff544430>",
+      "doc_to_target": "<function doc_to_target at 0x7f9fff544790>",
+      "doc_to_choice": "<function doc_to_choice at 0x7f9fff544af0>",
+      "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"
+    }
+  },
+  "versions": {
+    "arc_challenge": "Yaml",
+    "arc_easy": "Yaml",
+    "boolq": "Yaml",
+    "hellaswag": "Yaml",
+    "lambada_openai": "Yaml",
+    "openbookqa": "Yaml",
+    "piqa": "Yaml",
+    "sciq": "Yaml",
+    "wikitext": "Yaml",
+    "winogrande": "Yaml"
+  },
+  "config": {
+    "model": "hf",
+    "model_args": "pretrained=lomahony/eleuther-pythia12b-hh-sft,parallelize=True",
+    "batch_size": "16",
+    "batch_sizes": [],
+    "device": null,
+    "use_cache": null,
+    "limit": null,
+    "bootstrap_iters": 100000
+  },
+  "git_hash": "4cda3a1c"
+}

sft-pythia-12b-5shot-shelloutput.txt

@@ -0,0 +1,24 @@
+bootstrapping for stddev: perplexity
+hf (pretrained=lomahony/eleuther-pythia12b-hh-sft,parallelize=True), limit: None, num_fewshot: 5, batch_size: 8
+|     Task     |Version|Filter|    Metric     | Value |   |Stderr|
+|--------------|-------|------|---------------|------:|---|-----:|
+|arc_challenge |Yaml   |none  |acc            | 0.3712|±  |0.0141|
+|              |       |none  |acc_norm       | 0.4036|±  |0.0143|
+|arc_easy      |Yaml   |none  |acc            | 0.7201|±  |0.0092|
+|              |       |none  |acc_norm       | 0.7273|±  |0.0091|
+|boolq         |Yaml   |none  |acc            | 0.7220|±  |0.0078|
+|hellaswag     |Yaml   |none  |acc            | 0.5075|±  |0.0050|
+|              |       |none  |acc_norm       | 0.6749|±  |0.0047|
+|lambada_openai|Yaml   |none  |perplexity     | 4.3599|±  |0.1008|
+|              |       |none  |acc            | 0.6697|±  |0.0066|
+|openbookqa    |Yaml   |none  |acc            | 0.3080|±  |0.0207|
+|              |       |none  |acc_norm       | 0.4040|±  |0.0220|
+|piqa          |Yaml   |none  |acc            | 0.7677|±  |0.0099|
+|              |       |none  |acc_norm       | 0.7704|±  |0.0098|
+|sciq          |Yaml   |none  |acc            | 0.9450|±  |0.0072|
+|              |       |none  |acc_norm       | 0.9480|±  |0.0070|
+|wikitext      |Yaml   |none  |word_perplexity|16.5611|   |      |
+|              |       |none  |byte_perplexity| 1.5884|   |      |
+|              |       |none  |bits_per_byte  | 0.6676|   |      |
+|winogrande    |Yaml   |none  |acc            | 0.6346|±  |0.0135|
+

sft-pythia-12b-5shot/results.json

@@ -0,0 +1,388 @@
+{
+  "results": {
+    "arc_challenge": {
+      "acc,none": 0.371160409556314,
+      "acc_stderr,none": 0.014117971901142818,
+      "acc_norm,none": 0.4035836177474403,
+      "acc_norm_stderr,none": 0.014337158914268436
+    },
+    "arc_easy": {
+      "acc,none": 0.7201178451178452,
+      "acc_stderr,none": 0.009212077524656533,
+      "acc_norm,none": 0.7272727272727273,
+      "acc_norm_stderr,none": 0.00913863072636423
+    },
+    "boolq": {
+      "acc,none": 0.7220183486238532,
+      "acc_stderr,none": 0.00783564446741566
+    },
+    "hellaswag": {
+      "acc,none": 0.507468631746664,
+      "acc_stderr,none": 0.004989224715784542,
+      "acc_norm,none": 0.6748655646285601,
+      "acc_norm_stderr,none": 0.004674677287148605
+    },
+    "lambada_openai": {
+      "perplexity,none": 4.359872504756906,
+      "perplexity_stderr,none": 0.10082046734843834,
+      "acc,none": 0.6697069668154473,
+      "acc_stderr,none": 0.006552457124918184
+    },
+    "openbookqa": {
+      "acc,none": 0.308,
+      "acc_stderr,none": 0.0206670329874661,
+      "acc_norm,none": 0.404,
+      "acc_norm_stderr,none": 0.02196663529383293
+    },
+    "piqa": {
+      "acc,none": 0.7676822633297062,
+      "acc_stderr,none": 0.009853201384168241,
+      "acc_norm,none": 0.7704026115342764,
+      "acc_norm_stderr,none": 0.009812682950815206
+    },
+    "sciq": {
+      "acc,none": 0.945,
+      "acc_stderr,none": 0.007212976294639239,
+      "acc_norm,none": 0.948,
+      "acc_norm_stderr,none": 0.007024624213817149
+    },
+    "wikitext": {
+      "word_perplexity,none": 16.56112447245529,
+      "byte_perplexity,none": 1.588448103407034,
+      "bits_per_byte,none": 0.6676179561638627
+    },
+    "winogrande": {
+      "acc,none": 0.6345698500394633,
+      "acc_stderr,none": 0.013533965097638798
+    }
+  },
+  "configs": {
+    "arc_challenge": {
+      "task": "arc_challenge",
+      "group": [
+        "ai2_arc"
+      ],
+      "dataset_path": "ai2_arc",
+      "dataset_name": "ARC-Challenge",
+      "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": 5,
+      "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:"
+    },
+    "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": 5,
+      "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:"
+    },
+    "boolq": {
+      "task": "boolq",
+      "group": [
+        "super-glue-lm-eval-v1"
+      ],
+      "dataset_path": "super_glue",
+      "dataset_name": "boolq",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "{{passage}}\nQuestion: {{question}}?\nAnswer:",
+      "doc_to_target": "label",
+      "doc_to_choice": [
+        "no",
+        "yes"
+      ],
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "metric_list": [
+        {
+          "metric": "acc"
+        }
+      ],
+      "output_type": "multiple_choice",
+      "repeats": 1,
+      "should_decontaminate": true,
+      "doc_to_decontamination_query": "passage"
+    },
+    "hellaswag": {
+      "task": "hellaswag",
+      "group": [
+        "multiple_choice"
+      ],
+      "dataset_path": "hellaswag",
+      "training_split": "train",
+      "validation_split": "validation",
+      "process_docs": "<function process_docs at 0x7fb4b75e5700>",
+      "doc_to_text": "{{query}}",
+      "doc_to_target": "{{label}}",
+      "doc_to_choice": "{{choices}}",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "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": false
+    },
+    "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": 5,
+      "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}}"
+    },
+    "openbookqa": {
+      "task": "openbookqa",
+      "dataset_path": "openbookqa",
+      "dataset_name": "main",
+      "training_split": "train",
+      "validation_split": "validation",
+      "test_split": "test",
+      "doc_to_text": "question_stem",
+      "doc_to_target": "{{choices.label.index(answerKey.lstrip())}}",
+      "doc_to_choice": "{{choices.text}}",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "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_stem"
+    },
+    "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": 5,
+      "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"
+    },
+    "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": 5,
+      "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}}"
+    },
+    "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 0x7fb4b654f310>",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "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}}"
+    },
+    "winogrande": {
+      "task": "winogrande",
+      "dataset_path": "winogrande",
+      "dataset_name": "winogrande_xl",
+      "training_split": "train",
+      "validation_split": "validation",
+      "doc_to_text": "<function doc_to_text at 0x7fb4b654f550>",
+      "doc_to_target": "<function doc_to_target at 0x7fb4b654f8b0>",
+      "doc_to_choice": "<function doc_to_choice at 0x7fb4b654fc10>",
+      "description": "",
+      "target_delimiter": " ",
+      "fewshot_delimiter": "\n\n",
+      "num_fewshot": 5,
+      "metric_list": [
+        {
+          "metric": "acc",
+          "aggregation": "mean",
+          "higher_is_better": true
+        }
+      ],
+      "output_type": "multiple_choice",
+      "repeats": 1,
+      "should_decontaminate": true,
+      "doc_to_decontamination_query": "sentence"
+    }
+  },
+  "versions": {
+    "arc_challenge": "Yaml",
+    "arc_easy": "Yaml",
+    "boolq": "Yaml",
+    "hellaswag": "Yaml",
+    "lambada_openai": "Yaml",
+    "openbookqa": "Yaml",
+    "piqa": "Yaml",
+    "sciq": "Yaml",
+    "wikitext": "Yaml",
+    "winogrande": "Yaml"
+  },
+  "config": {
+    "model": "hf",
+    "model_args": "pretrained=lomahony/eleuther-pythia12b-hh-sft,parallelize=True",
+    "batch_size": "8",
+    "batch_sizes": [],
+    "device": null,
+    "use_cache": null,
+    "limit": null,
+    "bootstrap_iters": 100000
+  },
+  "git_hash": "4cda3a1c"
+}