lomahony/eleuther-pythia2.8b-hh-sft
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: lomahony/eleuther-pythia2.8b-hh-sft
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-01 12:43:18 +08:00
commit 05813afb02
19 changed files with 102432 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
sft-2.8b-eval-files/EleutherAI-pythia-2.8b-0shot-shelloutput.txt Normal file
View File

@@ -0,0 +1,24 @@
bootstrapping for stddev: perplexity
hf (pretrained=EleutherAI/pythia-2.8b), limit: None, num_fewshot: 0, batch_size: 16
| Task |Version|Filter| Metric | Value | |Stderr|
|--------------|-------|------|---------------|------:|---|-----:|
|arc_challenge |Yaml |none |acc | 0.2952|± |0.0133|
| | |none |acc_norm | 0.3311|± |0.0138|
|arc_easy |Yaml |none |acc | 0.6448|± |0.0098|
| | |none |acc_norm | 0.5871|± |0.0101|
|boolq |Yaml |none |acc | 0.6474|± |0.0084|
|hellaswag |Yaml |none |acc | 0.4532|± |0.0050|
| | |none |acc_norm | 0.5926|± |0.0049|
|lambada_openai|Yaml |none |perplexity | 5.0370|± |0.1191|
| | |none |acc | 0.6472|± |0.0067|
|openbookqa |Yaml |none |acc | 0.2400|± |0.0191|
| | |none |acc_norm | 0.3580|± |0.0215|
|piqa |Yaml |none |acc | 0.7394|± |0.0102|
| | |none |acc_norm | 0.7356|± |0.0103|
|sciq |Yaml |none |acc | 0.8870|± |0.0100|
| | |none |acc_norm | 0.8350|± |0.0117|
|wikitext |Yaml |none |word_perplexity|20.0571| | |
| | |none |byte_perplexity| 1.6394| | |
| | |none |bits_per_byte | 0.7132| | |
|winogrande |Yaml |none |acc | 0.5975|± |0.0138|

24
sft-2.8b-eval-files/EleutherAI-pythia-2.8b-5shot-shelloutput.txt Normal file
View File

@@ -0,0 +1,24 @@
bootstrapping for stddev: perplexity
hf (pretrained=EleutherAI/pythia-2.8b), limit: None, num_fewshot: 5, batch_size: 8
| Task |Version|Filter| Metric | Value | |Stderr|
|--------------|-------|------|---------------|------:|---|-----:|
|arc_challenge |Yaml |none |acc | 0.3123|± |0.0135|
| | |none |acc_norm | 0.3549|± |0.0140|
|arc_easy |Yaml |none |acc | 0.6700|± |0.0096|
| | |none |acc_norm | 0.6759|± |0.0096|
|boolq |Yaml |none |acc | 0.6664|± |0.0082|
|hellaswag |Yaml |none |acc | 0.4504|± |0.0050|
| | |none |acc_norm | 0.6031|± |0.0049|
|lambada_openai|Yaml |none |perplexity | 6.4479|± |0.1602|
| | |none |acc | 0.5948|± |0.0068|
|openbookqa |Yaml |none |acc | 0.2480|± |0.0193|
| | |none |acc_norm | 0.3720|± |0.0216|
|piqa |Yaml |none |acc | 0.7427|± |0.0102|
| | |none |acc_norm | 0.7443|± |0.0102|
|sciq |Yaml |none |acc | 0.9440|± |0.0073|
| | |none |acc_norm | 0.9400|± |0.0075|
|wikitext |Yaml |none |word_perplexity|20.0571| | |
| | |none |byte_perplexity| 1.6394| | |
| | |none |bits_per_byte | 0.7132| | |
|winogrande |Yaml |none |acc | 0.5943|± |0.0138|

24
sft-2.8b-eval-files/dpo-pythia-2.8b-0shot-shelloutput.txt Normal file
View File

@@ -0,0 +1,24 @@
bootstrapping for stddev: perplexity
hf (pretrained=lomahony/eleuther-pythia2.8b-hh-dpo), limit: None, num_fewshot: 0, batch_size: 16
| Task |Version|Filter| Metric | Value | |Stderr|
|--------------|-------|------|---------------|------:|---|-----:|
|arc_challenge |Yaml |none |acc | 0.3302|± |0.0137|
| | |none |acc_norm | 0.3490|± |0.0139|
|arc_easy |Yaml |none |acc | 0.6625|± |0.0097|
| | |none |acc_norm | 0.5918|± |0.0101|
|boolq |Yaml |none |acc | 0.6248|± |0.0085|
|hellaswag |Yaml |none |acc | 0.4677|± |0.0050|
| | |none |acc_norm | 0.6072|± |0.0049|
|lambada_openai|Yaml |none |perplexity | 4.4821|± |0.1220|
| | |none |acc | 0.6350|± |0.0067|
|openbookqa |Yaml |none |acc | 0.2640|± |0.0197|
| | |none |acc_norm | 0.3960|± |0.0219|
|piqa |Yaml |none |acc | 0.7535|± |0.0101|
| | |none |acc_norm | 0.7454|± |0.0102|
|sciq |Yaml |none |acc | 0.8630|± |0.0109|
| | |none |acc_norm | 0.8030|± |0.0126|
|wikitext |Yaml |none |word_perplexity|21.9279| | |
| | |none |byte_perplexity| 1.6637| | |
| | |none |bits_per_byte | 0.7344| | |
|winogrande |Yaml |none |acc | 0.5967|± |0.0138|

24
sft-2.8b-eval-files/dpo-pythia-2.8b-5shot-shelloutput.txt Normal file
View File

@@ -0,0 +1,24 @@
bootstrapping for stddev: perplexity
hf (pretrained=lomahony/eleuther-pythia2.8b-hh-dpo), limit: None, num_fewshot: 5, batch_size: 8
| Task |Version|Filter| Metric | Value | |Stderr|
|--------------|-------|------|---------------|------:|---|-----:|
|arc_challenge |Yaml |none |acc | 0.3353|± |0.0138|
| | |none |acc_norm | 0.3788|± |0.0142|
|arc_easy |Yaml |none |acc | 0.6890|± |0.0095|
| | |none |acc_norm | 0.6936|± |0.0095|
|boolq |Yaml |none |acc | 0.6495|± |0.0083|
|hellaswag |Yaml |none |acc | 0.4680|± |0.0050|
| | |none |acc_norm | 0.6124|± |0.0049|
|lambada_openai|Yaml |none |perplexity | 5.9966|± |0.1685|
| | |none |acc | 0.5789|± |0.0069|
|openbookqa |Yaml |none |acc | 0.2720|± |0.0199|
| | |none |acc_norm | 0.3740|± |0.0217|
|piqa |Yaml |none |acc | 0.7535|± |0.0101|
| | |none |acc_norm | 0.7573|± |0.0100|
|sciq |Yaml |none |acc | 0.9340|± |0.0079|
| | |none |acc_norm | 0.9190|± |0.0086|
|wikitext |Yaml |none |word_perplexity|21.9279| | |
| | |none |byte_perplexity| 1.6637| | |
| | |none |bits_per_byte | 0.7344| | |
|winogrande |Yaml |none |acc | 0.5967|± |0.0138|

24
sft-2.8b-eval-files/sft-pythia-2.8b-0shot-shelloutput.txt Normal file
View File

@@ -0,0 +1,24 @@
bootstrapping for stddev: perplexity
hf (pretrained=lomahony/eleuther-pythia2.8b-hh-sft), limit: None, num_fewshot: 0, batch_size: 16
| Task |Version|Filter| Metric | Value | |Stderr|
|--------------|-------|------|---------------|------:|---|-----:|
|arc_challenge |Yaml |none |acc | 0.3003|± |0.0134|
| | |none |acc_norm | 0.3268|± |0.0137|
|arc_easy |Yaml |none |acc | 0.6486|± |0.0098|
| | |none |acc_norm | 0.5657|± |0.0102|
|boolq |Yaml |none |acc | 0.6468|± |0.0084|
|hellaswag |Yaml |none |acc | 0.4516|± |0.0050|
| | |none |acc_norm | 0.5870|± |0.0049|
|lambada_openai|Yaml |none |perplexity | 4.9120|± |0.1230|
| | |none |acc | 0.6344|± |0.0067|
|openbookqa |Yaml |none |acc | 0.2540|± |0.0195|
| | |none |acc_norm | 0.3700|± |0.0216|
|piqa |Yaml |none |acc | 0.7448|± |0.0102|
| | |none |acc_norm | 0.7405|± |0.0102|
|sciq |Yaml |none |acc | 0.8720|± |0.0106|
| | |none |acc_norm | 0.8010|± |0.0126|
|wikitext |Yaml |none |word_perplexity|20.7220| | |
| | |none |byte_perplexity| 1.6482| | |
| | |none |bits_per_byte | 0.7209| | |
|winogrande |Yaml |none |acc | 0.5888|± |0.0138|

404
sft-2.8b-eval-files/sft-pythia-2.8b-0shot/results.json Normal file
View File

@@ -0,0 +1,404 @@
{
"results": {
"arc_challenge": {
"acc,none": 0.3003412969283277,
"acc_stderr,none": 0.013395909309957,
"acc_norm,none": 0.3267918088737201,
"acc_norm_stderr,none": 0.013706665975587336
},
"arc_easy": {
"acc,none": 0.6485690235690236,
"acc_stderr,none": 0.00979639558281772,
"acc_norm,none": 0.5656565656565656,
"acc_norm_stderr,none": 0.010170943451269428
},
"boolq": {
"acc,none": 0.6467889908256881,
"acc_stderr,none": 0.008359705247064303
},
"hellaswag": {
"acc,none": 0.45160326628161723,
"acc_stderr,none": 0.004966351835028203,
"acc_norm,none": 0.5870344552877913,
"acc_norm_stderr,none": 0.004913604782665858
},
"lambada_openai": {
"perplexity,none": 4.911988887150992,
"perplexity_stderr,none": 0.12302146370246485,
"acc,none": 0.6343877352998254,
"acc_stderr,none": 0.006709649590864073
},
"openbookqa": {
"acc,none": 0.254,
"acc_stderr,none": 0.01948659680164338,
"acc_norm,none": 0.37,
"acc_norm_stderr,none": 0.021613289165165785
},
"piqa": {
"acc,none": 0.7448313384113167,
"acc_stderr,none": 0.010171571592521822,
"acc_norm,none": 0.7404787812840044,
"acc_norm_stderr,none": 0.01022793988817392
},
"sciq": {
"acc,none": 0.872,
"acc_stderr,none": 0.010570133761108656,
"acc_norm,none": 0.801,
"acc_norm_stderr,none": 0.012631649083099182
},
"wikitext": {
"word_perplexity,none": 20.722018295590285,
"byte_perplexity,none": 1.6482397447214083,
"bits_per_byte,none": 0.720926104999623
},
"winogrande": {
"acc,none": 0.5887924230465666,
"acc_stderr,none": 0.013829128358676862
}
},
"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 0x7faa07711120>",
"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 0x7faa07711360>",
"doc_to_target": "<function doc_to_target at 0x7faa077116c0>",
"doc_to_choice": "<function doc_to_choice at 0x7faa07711a20>",
"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-pythia2.8b-hh-sft",
"batch_size": "16",
"batch_sizes": [],
"device": null,
"use_cache": null,
"limit": null,
"bootstrap_iters": 100000
},
"git_hash": "d1a44c8"
}

24
sft-2.8b-eval-files/sft-pythia-2.8b-5shot-shelloutput.txt Normal file
View File

@@ -0,0 +1,24 @@
bootstrapping for stddev: perplexity
hf (pretrained=lomahony/eleuther-pythia2.8b-hh-sft), limit: None, num_fewshot: 5, batch_size: 8
| Task |Version|Filter| Metric | Value | |Stderr|
|--------------|-------|------|---------------|------:|---|-----:|
|arc_challenge |Yaml |none |acc | 0.3319|± |0.0138|
| | |none |acc_norm | 0.3584|± |0.0140|
|arc_easy |Yaml |none |acc | 0.6839|± |0.0095|
| | |none |acc_norm | 0.6843|± |0.0095|
|boolq |Yaml |none |acc | 0.6700|± |0.0082|
|hellaswag |Yaml |none |acc | 0.4509|± |0.0050|
| | |none |acc_norm | 0.5901|± |0.0049|
|lambada_openai|Yaml |none |perplexity | 6.5403|± |0.1706|
| | |none |acc | 0.5758|± |0.0069|
|openbookqa |Yaml |none |acc | 0.2560|± |0.0195|
| | |none |acc_norm | 0.3580|± |0.0215|
|piqa |Yaml |none |acc | 0.7486|± |0.0101|
| | |none |acc_norm | 0.7421|± |0.0102|
|sciq |Yaml |none |acc | 0.9410|± |0.0075|
| | |none |acc_norm | 0.9370|± |0.0077|
|wikitext |Yaml |none |word_perplexity|20.7220| | |
| | |none |byte_perplexity| 1.6482| | |
| | |none |bits_per_byte | 0.7209| | |
|winogrande |Yaml |none |acc | 0.5904|± |0.0138|

404
sft-2.8b-eval-files/sft-pythia-2.8b-5shot/results.json Normal file
View File

@@ -0,0 +1,404 @@
{
"results": {
"arc_challenge": {
"acc,none": 0.3319112627986348,
"acc_stderr,none": 0.013760988200880536,
"acc_norm,none": 0.3583617747440273,
"acc_norm_stderr,none": 0.014012883334859862
},
"arc_easy": {
"acc,none": 0.6839225589225589,
"acc_stderr,none": 0.009540440071928283,
"acc_norm,none": 0.6843434343434344,
"acc_norm_stderr,none": 0.009537019245566084
},
"boolq": {
"acc,none": 0.6700305810397553,
"acc_stderr,none": 0.008223878741654847
},
"hellaswag": {
"acc,none": 0.4509061939852619,
"acc_stderr,none": 0.004965670398127348,
"acc_norm,none": 0.5901214897430791,
"acc_norm_stderr,none": 0.004908059353503842
},
"lambada_openai": {
"perplexity,none": 6.540250350436668,
"perplexity_stderr,none": 0.17057269700844382,
"acc,none": 0.5757810983892878,
"acc_stderr,none": 0.006885504751619328
},
"openbookqa": {
"acc,none": 0.256,
"acc_stderr,none": 0.019536923574747605,
"acc_norm,none": 0.358,
"acc_norm_stderr,none": 0.021461434862859115
},
"piqa": {
"acc,none": 0.7486398258977149,
"acc_stderr,none": 0.010121156016819257,
"acc_norm,none": 0.7421109902067464,
"acc_norm_stderr,none": 0.010206956662056257
},
"sciq": {
"acc,none": 0.941,
"acc_stderr,none": 0.007454835650406723,
"acc_norm,none": 0.937,
"acc_norm_stderr,none": 0.007687007876286444
},
"wikitext": {
"word_perplexity,none": 20.722018295590285,
"byte_perplexity,none": 1.6482397447214083,
"bits_per_byte,none": 0.720926104999623
},
"winogrande": {
"acc,none": 0.590370955011839,
"acc_stderr,none": 0.013821049109655478
}
},
"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 0x7fbe1b3e5120>",
"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 0x7fbe1b3e5360>",
"doc_to_target": "<function doc_to_target at 0x7fbe1b3e56c0>",
"doc_to_choice": "<function doc_to_choice at 0x7fbe1b3e5a20>",
"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-pythia2.8b-hh-sft",
"batch_size": "8",
"batch_sizes": [],
"device": null,
"use_cache": null,
"limit": null,
"bootstrap_iters": 100000
},
"git_hash": "d1a44c8"
}