Upload 8 files

base-12b-eval-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|
+

base-12b-eval-files/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"
+}

base-12b-eval-files/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|
+

base-12b-eval-files/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-12b-eval-files/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: 4
+|     Task     |Version|Filter|    Metric     | Value |   |Stderr|
+|--------------|-------|------|---------------|------:|---|-----:|
+|arc_challenge |Yaml   |none  |acc            | 0.3131|±  |0.0136|
+|              |       |none  |acc_norm       | 0.3447|±  |0.0139|
+|arc_easy      |Yaml   |none  |acc            | 0.7041|±  |0.0094|
+|              |       |none  |acc_norm       | 0.6178|±  |0.0100|
+|boolq         |Yaml   |none  |acc            | 0.6945|±  |0.0081|
+|hellaswag     |Yaml   |none  |acc            | 0.5056|±  |0.0050|
+|              |       |none  |acc_norm       | 0.6685|±  |0.0047|
+|lambada_openai|Yaml   |none  |perplexity     | 3.5396|±  |0.0776|
+|              |       |none  |acc            | 0.7093|±  |0.0063|
+|openbookqa    |Yaml   |none  |acc            | 0.2760|±  |0.0200|
+|              |       |none  |acc_norm       | 0.3820|±  |0.0218|
+|piqa          |Yaml   |none  |acc            | 0.7655|±  |0.0099|
+|              |       |none  |acc_norm       | 0.7655|±  |0.0099|
+|sciq          |Yaml   |none  |acc            | 0.9060|±  |0.0092|
+|              |       |none  |acc_norm       | 0.8410|±  |0.0116|
+|wikitext      |Yaml   |none  |word_perplexity|16.4972|   |      |
+|              |       |none  |byte_perplexity| 1.5874|   |      |
+|              |       |none  |bits_per_byte  | 0.6667|   |      |
+|winogrande    |Yaml   |none  |acc            | 0.6354|±  |0.0135|
+

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

@@ -0,0 +1,404 @@
+{
+  "results": {
+    "arc_challenge": {
+      "acc,none": 0.31313993174061433,
+      "acc_stderr,none": 0.013552671543623489,
+      "acc_norm,none": 0.3447098976109215,
+      "acc_norm_stderr,none": 0.013888816286782114
+    },
+    "arc_easy": {
+      "acc,none": 0.7041245791245792,
+      "acc_stderr,none": 0.009365854134140067,
+      "acc_norm,none": 0.6178451178451179,
+      "acc_norm_stderr,none": 0.00997074728129243
+    },
+    "boolq": {
+      "acc,none": 0.6944954128440367,
+      "acc_stderr,none": 0.00805630868516481
+    },
+    "hellaswag": {
+      "acc,none": 0.5055765783708425,
+      "acc_stderr,none": 0.0049894710550909506,
+      "acc_norm,none": 0.6684923322047401,
+      "acc_norm_stderr,none": 0.004697929774670318
+    },
+    "lambada_openai": {
+      "perplexity,none": 3.5396245492779395,
+      "perplexity_stderr,none": 0.07756934951997978,
+      "acc,none": 0.7092955559868038,
+      "acc_stderr,none": 0.006326325417865827
+    },
+    "openbookqa": {
+      "acc,none": 0.276,
+      "acc_stderr,none": 0.02001121929807353,
+      "acc_norm,none": 0.382,
+      "acc_norm_stderr,none": 0.021750820591250827
+    },
+    "piqa": {
+      "acc,none": 0.76550598476605,
+      "acc_stderr,none": 0.009885203143240548,
+      "acc_norm,none": 0.76550598476605,
+      "acc_norm_stderr,none": 0.009885203143240548
+    },
+    "sciq": {
+      "acc,none": 0.906,
+      "acc_stderr,none": 0.00923305200078773,
+      "acc_norm,none": 0.841,
+      "acc_norm_stderr,none": 0.01156947936827129
+    },
+    "wikitext": {
+      "word_perplexity,none": 16.49715176032047,
+      "byte_perplexity,none": 1.5874349349461754,
+      "bits_per_byte,none": 0.6666974605750762
+    },
+    "winogrande": {
+      "acc,none": 0.6353591160220995,
+      "acc_stderr,none": 0.01352774662242984
+    }
+  },
+  "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 0x7f91f8185120>",
+      "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 0x7f91f8185360>",
+      "doc_to_target": "<function doc_to_target at 0x7f91f81856c0>",
+      "doc_to_choice": "<function doc_to_choice at 0x7f91f8185a20>",
+      "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": "4",
+    "batch_sizes": [],
+    "device": null,
+    "use_cache": null,
+    "limit": null,
+    "bootstrap_iters": 100000
+  },
+  "git_hash": "d1a44c8"
+}

sft-12b-eval-files/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: 4
+|     Task     |Version|Filter|    Metric     | Value |   |Stderr|
+|--------------|-------|------|---------------|------:|---|-----:|
+|arc_challenge |Yaml   |none  |acc            | 0.3712|±  |0.0141|
+|              |       |none  |acc_norm       | 0.4010|±  |0.0143|
+|arc_easy      |Yaml   |none  |acc            | 0.7189|±  |0.0092|
+|              |       |none  |acc_norm       | 0.7285|±  |0.0091|
+|boolq         |Yaml   |none  |acc            | 0.7174|±  |0.0079|
+|hellaswag     |Yaml   |none  |acc            | 0.5076|±  |0.0050|
+|              |       |none  |acc_norm       | 0.6764|±  |0.0047|
+|lambada_openai|Yaml   |none  |perplexity     | 4.3625|±  |0.1008|
+|              |       |none  |acc            | 0.6703|±  |0.0065|
+|openbookqa    |Yaml   |none  |acc            | 0.3120|±  |0.0207|
+|              |       |none  |acc_norm       | 0.4000|±  |0.0219|
+|piqa          |Yaml   |none  |acc            | 0.7688|±  |0.0098|
+|              |       |none  |acc_norm       | 0.7704|±  |0.0098|
+|sciq          |Yaml   |none  |acc            | 0.9470|±  |0.0071|
+|              |       |none  |acc_norm       | 0.9500|±  |0.0069|
+|wikitext      |Yaml   |none  |word_perplexity|16.4972|   |      |
+|              |       |none  |byte_perplexity| 1.5874|   |      |
+|              |       |none  |bits_per_byte  | 0.6667|   |      |
+|winogrande    |Yaml   |none  |acc            | 0.6354|±  |0.0135|
+

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

@@ -0,0 +1,404 @@
+{
+  "results": {
+    "arc_challenge": {
+      "acc,none": 0.371160409556314,
+      "acc_stderr,none": 0.014117971901142818,
+      "acc_norm,none": 0.40102389078498296,
+      "acc_norm_stderr,none": 0.014322255790719867
+    },
+    "arc_easy": {
+      "acc,none": 0.7188552188552189,
+      "acc_stderr,none": 0.009224735470286998,
+      "acc_norm,none": 0.7285353535353535,
+      "acc_norm_stderr,none": 0.009125362970360623
+    },
+    "boolq": {
+      "acc,none": 0.7174311926605504,
+      "acc_stderr,none": 0.007874895085575213
+    },
+    "hellaswag": {
+      "acc,none": 0.5075682135032862,
+      "acc_stderr,none": 0.004989209770743232,
+      "acc_norm,none": 0.6763592909778928,
+      "acc_norm_stderr,none": 0.004669085411342217
+    },
+    "lambada_openai": {
+      "perplexity,none": 4.362525202130014,
+      "perplexity_stderr,none": 0.10078560700527144,
+      "acc,none": 0.6702891519503202,
+      "acc_stderr,none": 0.006549524731584282
+    },
+    "openbookqa": {
+      "acc,none": 0.312,
+      "acc_stderr,none": 0.020740596536488073,
+      "acc_norm,none": 0.4,
+      "acc_norm_stderr,none": 0.021930844120728505
+    },
+    "piqa": {
+      "acc,none": 0.7687704026115343,
+      "acc_stderr,none": 0.009837063180625326,
+      "acc_norm,none": 0.7704026115342764,
+      "acc_norm_stderr,none": 0.009812682950815206
+    },
+    "sciq": {
+      "acc,none": 0.947,
+      "acc_stderr,none": 0.007088105617246442,
+      "acc_norm,none": 0.95,
+      "acc_norm_stderr,none": 0.006895472974897893
+    },
+    "wikitext": {
+      "word_perplexity,none": 16.49715176032047,
+      "byte_perplexity,none": 1.5874349349461754,
+      "bits_per_byte,none": 0.6666974605750762
+    },
+    "winogrande": {
+      "acc,none": 0.6353591160220995,
+      "acc_stderr,none": 0.01352774662242984
+    }
+  },
+  "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 0x7f05d3475120>",
+      "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 0x7f05d3475360>",
+      "doc_to_target": "<function doc_to_target at 0x7f05d34756c0>",
+      "doc_to_choice": "<function doc_to_choice at 0x7f05d3475a20>",
+      "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": "4",
+    "batch_sizes": [],
+    "device": null,
+    "use_cache": null,
+    "limit": null,
+    "bootstrap_iters": 100000
+  },
+  "git_hash": "d1a44c8"
+}