|
---
|
|
license: apache-2.0
|
|
pipeline_tag: text-generation
|
|
library_name: transformers
|
|
language: [
|
|
'en', 'am', 'ar', 'as', 'az', 'be', 'bg', 'bn', 'br', 'bs', 'ca', 'cs', 'cy', 'da', 'de', 'el',
|
|
'eo', 'es', 'et', 'eu', 'fa', 'ff', 'fi', 'fr', 'fy', 'ga', 'gd', 'gl', 'gn', 'gu', 'ha', 'he',
|
|
'hi', 'hr', 'ht', 'hu', 'hy', 'id', 'ig', 'is', 'it', 'ja', 'jv', 'ka', 'kk', 'km', 'kn', 'ko',
|
|
'ku', 'ky', 'la', 'lg', 'li', 'ln', 'lo', 'lt', 'lv', 'mg', 'mk', 'ml', 'mn', 'mr', 'ms', 'my',
|
|
'ne', 'nl', 'no', 'ns', 'om', 'or', 'pa', 'pl', 'ps', 'pt', 'qu', 'rm', 'ro', 'ru', 'sa', 'si',
|
|
'sc', 'sd', 'sk', 'sl', 'so', 'sq', 'sr', 'ss', 'su', 'sv', 'sw', 'ta', 'te', 'th', 'tl', 'tn',
|
|
'tr', 'ug', 'uk', 'ur', 'uz', 'vi', 'wo', 'xh', 'yi', 'yo', 'zu',
|
|
]
|
|
datasets: [
|
|
'xu-song/cc100-samples',
|
|
'jordiclive/wikipedia-summary-dataset',
|
|
'JeanKaddour/minipile',
|
|
'badrex/llm-emoji-dataset',
|
|
'fblgit/simple-math',
|
|
'Gusarich/math-expressions-1m',
|
|
'AtlasUnified/atlas-math-sets',
|
|
'gair-prox/open-web-math-pro',
|
|
'bigcode/the-stack-smol-xs',
|
|
'rombodawg/code_bagel',
|
|
'AtlasUnified/Atlas-Reasoning',
|
|
'thesven/gsm8k-reasoning',
|
|
'AlgorithmicResearchGroup/math_reasoning_autoformalization_track',
|
|
'KingNish/reasoning-base-20k',
|
|
'SkunkworksAI/reasoning-0.01',
|
|
'Magpie-Align/Magpie-Reasoning-150K',
|
|
]
|
|
tags:
|
|
- litgpt
|
|
- litdata
|
|
---
|
|
|
|
# tangled-llama-b-128k-base-v0.1
|
|
|
|
|
|
|
|
A pretrained language model based on the Llama model with about **62.9M** parameters. This model has been trained on **10.6B** (`10,630,121,844`) tokens from more than **31.3M** (`31,383,840`) dataset rows.
|
|
|
|
This model **isn't** designed for immediate use but rather for Continued Pretraining and Finetuning on a downstream task. While it can handle a context length of up to **128K** (`131,072`) tokens, it was pretrained with sequences of **2K** (`2048`) tokens.
|
|
|
|
The objective is to streamline the cognitive or reasoning core, eliminating any redundant knowledge from the model.
|
|
|
|
[loss, val_loss](https://api.wandb.ai/links/mtasic85/x4dxkpy7)
|
|
|
|
[val_ppl](https://api.wandb.ai/links/mtasic85/hr03vzo2)
|
|
|
|
[epoch](https://api.wandb.ai/links/mtasic85/e6ai0066)
|
|
|
|
[learning_rate](https://api.wandb.ai/links/mtasic85/lap8xl2w)
|
|
|
|
## Pretrain Evaluation
|
|
|
|
### lm-evaluation-harness
|
|
|
|
```bash
|
|
litgpt evaluate --tasks 'hellaswag,gsm8k,truthfulqa_mc2,mmlu,winogrande,arc_challenge' --out_dir 'evaluate-quick/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
|
|
```
|
|
|
|
| Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr|
|
|
|---------------------------------------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|
|
|arc_challenge | 1|none | 0|acc |↑ |0.1852|± |0.0114|
|
|
| | |none | 0|acc_norm |↑ |0.2201|± |0.0121|
|
|
|gsm8k | 3|flexible-extract| 5|exact_match|↑ |0.0205|± |0.0039|
|
|
| | |strict-match | 5|exact_match|↑ |0.0000|± |0.0000|
|
|
|hellaswag | 1|none | 0|acc |↑ |0.2628|± |0.0044|
|
|
| | |none | 0|acc_norm |↑ |0.2705|± |0.0044|
|
|
|mmlu | 2|none | |acc |↑ |0.2468|± |0.0036|
|
|
| - humanities | 2|none | |acc |↑ |0.2459|± |0.0063|
|
|
| - formal_logic | 1|none | 0|acc |↑ |0.3175|± |0.0416|
|
|
| - high_school_european_history | 1|none | 0|acc |↑ |0.2364|± |0.0332|
|
|
| - high_school_us_history | 1|none | 0|acc |↑ |0.2304|± |0.0296|
|
|
| - high_school_world_history | 1|none | 0|acc |↑ |0.2194|± |0.0269|
|
|
| - international_law | 1|none | 0|acc |↑ |0.2479|± |0.0394|
|
|
| - jurisprudence | 1|none | 0|acc |↑ |0.2315|± |0.0408|
|
|
| - logical_fallacies | 1|none | 0|acc |↑ |0.2147|± |0.0323|
|
|
| - moral_disputes | 1|none | 0|acc |↑ |0.2168|± |0.0222|
|
|
| - moral_scenarios | 1|none | 0|acc |↑ |0.2726|± |0.0149|
|
|
| - philosophy | 1|none | 0|acc |↑ |0.1865|± |0.0221|
|
|
| - prehistory | 1|none | 0|acc |↑ |0.2191|± |0.0230|
|
|
| - professional_law | 1|none | 0|acc |↑ |0.2490|± |0.0110|
|
|
| - world_religions | 1|none | 0|acc |↑ |0.3450|± |0.0365|
|
|
| - other | 2|none | |acc |↑ |0.2385|± |0.0076|
|
|
| - business_ethics | 1|none | 0|acc |↑ |0.2200|± |0.0416|
|
|
| - clinical_knowledge | 1|none | 0|acc |↑ |0.2264|± |0.0258|
|
|
| - college_medicine | 1|none | 0|acc |↑ |0.2601|± |0.0335|
|
|
| - global_facts | 1|none | 0|acc |↑ |0.1900|± |0.0394|
|
|
| - human_aging | 1|none | 0|acc |↑ |0.2422|± |0.0288|
|
|
| - management | 1|none | 0|acc |↑ |0.2330|± |0.0419|
|
|
| - marketing | 1|none | 0|acc |↑ |0.2821|± |0.0295|
|
|
| - medical_genetics | 1|none | 0|acc |↑ |0.2900|± |0.0456|
|
|
| - miscellaneous | 1|none | 0|acc |↑ |0.2388|± |0.0152|
|
|
| - nutrition | 1|none | 0|acc |↑ |0.1993|± |0.0229|
|
|
| - professional_accounting | 1|none | 0|acc |↑ |0.2270|± |0.0250|
|
|
| - professional_medicine | 1|none | 0|acc |↑ |0.2610|± |0.0267|
|
|
| - virology | 1|none | 0|acc |↑ |0.2349|± |0.0330|
|
|
| - social sciences | 2|none | |acc |↑ |0.2632|± |0.0079|
|
|
| - econometrics | 1|none | 0|acc |↑ |0.2544|± |0.0410|
|
|
| - high_school_geography | 1|none | 0|acc |↑ |0.1869|± |0.0278|
|
|
| - high_school_government_and_politics| 1|none | 0|acc |↑ |0.2850|± |0.0326|
|
|
| - high_school_macroeconomics | 1|none | 0|acc |↑ |0.3128|± |0.0235|
|
|
| - high_school_microeconomics | 1|none | 0|acc |↑ |0.2773|± |0.0291|
|
|
| - high_school_psychology | 1|none | 0|acc |↑ |0.2422|± |0.0184|
|
|
| - human_sexuality | 1|none | 0|acc |↑ |0.2595|± |0.0384|
|
|
| - professional_psychology | 1|none | 0|acc |↑ |0.2435|± |0.0174|
|
|
| - public_relations | 1|none | 0|acc |↑ |0.2273|± |0.0401|
|
|
| - security_studies | 1|none | 0|acc |↑ |0.3265|± |0.0300|
|
|
| - sociology | 1|none | 0|acc |↑ |0.2537|± |0.0308|
|
|
| - us_foreign_policy | 1|none | 0|acc |↑ |0.3000|± |0.0461|
|
|
| - stem | 2|none | |acc |↑ |0.2404|± |0.0076|
|
|
| - abstract_algebra | 1|none | 0|acc |↑ |0.1700|± |0.0378|
|
|
| - anatomy | 1|none | 0|acc |↑ |0.2074|± |0.0350|
|
|
| - astronomy | 1|none | 0|acc |↑ |0.2105|± |0.0332|
|
|
| - college_biology | 1|none | 0|acc |↑ |0.2153|± |0.0344|
|
|
| - college_chemistry | 1|none | 0|acc |↑ |0.2000|± |0.0402|
|
|
| - college_computer_science | 1|none | 0|acc |↑ |0.2300|± |0.0423|
|
|
| - college_mathematics | 1|none | 0|acc |↑ |0.1700|± |0.0378|
|
|
| - college_physics | 1|none | 0|acc |↑ |0.2647|± |0.0439|
|
|
| - computer_security | 1|none | 0|acc |↑ |0.2700|± |0.0446|
|
|
| - conceptual_physics | 1|none | 0|acc |↑ |0.2766|± |0.0292|
|
|
| - electrical_engineering | 1|none | 0|acc |↑ |0.2552|± |0.0363|
|
|
| - elementary_mathematics | 1|none | 0|acc |↑ |0.2566|± |0.0225|
|
|
| - high_school_biology | 1|none | 0|acc |↑ |0.2097|± |0.0232|
|
|
| - high_school_chemistry | 1|none | 0|acc |↑ |0.2611|± |0.0309|
|
|
| - high_school_computer_science | 1|none | 0|acc |↑ |0.2600|± |0.0441|
|
|
| - high_school_mathematics | 1|none | 0|acc |↑ |0.2111|± |0.0249|
|
|
| - high_school_physics | 1|none | 0|acc |↑ |0.2517|± |0.0354|
|
|
| - high_school_statistics | 1|none | 0|acc |↑ |0.3056|± |0.0314|
|
|
| - machine_learning | 1|none | 0|acc |↑ |0.2857|± |0.0429|
|
|
|truthfulqa_mc2 | 2|none | 0|acc |↑ |0.5010|± |0.0159|
|
|
|winogrande | 1|none | 0|acc |↑ |0.5130|± |0.0140|
|
|
|
|
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
|
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
|
|mmlu | 2|none | |acc |↑ |0.2468|± |0.0036|
|
|
| - humanities | 2|none | |acc |↑ |0.2459|± |0.0063|
|
|
| - other | 2|none | |acc |↑ |0.2385|± |0.0076|
|
|
| - social sciences| 2|none | |acc |↑ |0.2632|± |0.0079|
|
|
| - stem | 2|none | |acc |↑ |0.2404|± |0.0076|
|
|
|
|
```bash
|
|
litgpt evaluate --tasks 'leaderboard' --out_dir 'evaluate-leaderboard/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
|
|
```
|
|
|
|
| Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
|
|
|-----------------------------------------------------------|-------|------|-----:|-----------------------|---|-----:|---|------|
|
|
|leaderboard | N/A| | | | | | | |
|
|
| - leaderboard_bbh | N/A| | | | | | | |
|
|
| - leaderboard_bbh_boolean_expressions | 1|none | 3|acc_norm |↑ |0.4680|± |0.0316|
|
|
| - leaderboard_bbh_causal_judgement | 1|none | 3|acc_norm |↑ |0.5187|± |0.0366|
|
|
| - leaderboard_bbh_date_understanding | 1|none | 3|acc_norm |↑ |0.1880|± |0.0248|
|
|
| - leaderboard_bbh_disambiguation_qa | 1|none | 3|acc_norm |↑ |0.3440|± |0.0301|
|
|
| - leaderboard_bbh_formal_fallacies | 1|none | 3|acc_norm |↑ |0.4720|± |0.0316|
|
|
| - leaderboard_bbh_geometric_shapes | 1|none | 3|acc_norm |↑ |0.1200|± |0.0206|
|
|
| - leaderboard_bbh_hyperbaton | 1|none | 3|acc_norm |↑ |0.5240|± |0.0316|
|
|
| - leaderboard_bbh_logical_deduction_five_objects | 1|none | 3|acc_norm |↑ |0.2160|± |0.0261|
|
|
| - leaderboard_bbh_logical_deduction_seven_objects | 1|none | 3|acc_norm |↑ |0.1400|± |0.0220|
|
|
| - leaderboard_bbh_logical_deduction_three_objects | 1|none | 3|acc_norm |↑ |0.3200|± |0.0296|
|
|
| - leaderboard_bbh_movie_recommendation | 1|none | 3|acc_norm |↑ |0.2360|± |0.0269|
|
|
| - leaderboard_bbh_navigate | 1|none | 3|acc_norm |↑ |0.4200|± |0.0313|
|
|
| - leaderboard_bbh_object_counting | 1|none | 3|acc_norm |↑ |0.1000|± |0.0190|
|
|
| - leaderboard_bbh_penguins_in_a_table | 1|none | 3|acc_norm |↑ |0.1575|± |0.0303|
|
|
| - leaderboard_bbh_reasoning_about_colored_objects | 1|none | 3|acc_norm |↑ |0.0920|± |0.0183|
|
|
| - leaderboard_bbh_ruin_names | 1|none | 3|acc_norm |↑ |0.2480|± |0.0274|
|
|
| - leaderboard_bbh_salient_translation_error_detection | 1|none | 3|acc_norm |↑ |0.1200|± |0.0206|
|
|
| - leaderboard_bbh_snarks | 1|none | 3|acc_norm |↑ |0.4888|± |0.0376|
|
|
| - leaderboard_bbh_sports_understanding | 1|none | 3|acc_norm |↑ |0.4600|± |0.0316|
|
|
| - leaderboard_bbh_temporal_sequences | 1|none | 3|acc_norm |↑ |0.2440|± |0.0272|
|
|
| - leaderboard_bbh_tracking_shuffled_objects_five_objects | 1|none | 3|acc_norm |↑ |0.1560|± |0.0230|
|
|
| - leaderboard_bbh_tracking_shuffled_objects_seven_objects| 1|none | 3|acc_norm |↑ |0.0960|± |0.0187|
|
|
| - leaderboard_bbh_tracking_shuffled_objects_three_objects| 1|none | 3|acc_norm |↑ |0.3800|± |0.0308|
|
|
| - leaderboard_bbh_web_of_lies | 1|none | 3|acc_norm |↑ |0.4720|± |0.0316|
|
|
| - leaderboard_gpqa | N/A| | | | | | | |
|
|
| - leaderboard_gpqa_diamond | 1|none | 0|acc_norm |↑ |0.1970|± |0.0283|
|
|
| - leaderboard_gpqa_extended | 1|none | 0|acc_norm |↑ |0.2509|± |0.0186|
|
|
| - leaderboard_gpqa_main | 1|none | 0|acc_norm |↑ |0.2589|± |0.0207|
|
|
| - leaderboard_ifeval | 3|none | 0|inst_level_loose_acc |↑ |0.2650|± | N/A|
|
|
| | |none | 0|inst_level_strict_acc |↑ |0.2530|± | N/A|
|
|
| | |none | 0|prompt_level_loose_acc |↑ |0.1590|± |0.0157|
|
|
| | |none | 0|prompt_level_strict_acc|↑ |0.1553|± |0.0156|
|
|
| - leaderboard_math_hard | N/A| | | | | | | |
|
|
| - leaderboard_math_algebra_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_math_counting_and_prob_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_math_geometry_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_math_intermediate_algebra_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_math_num_theory_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_math_prealgebra_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_math_precalculus_hard | 1|none | 4|exact_match |↑ |0.0000|± | 0|
|
|
| - leaderboard_mmlu_pro | 0.1|none | 5|acc |↑ |0.1174|± |0.0029|
|
|
| - leaderboard_musr | N/A| | | | | | | |
|
|
| - leaderboard_musr_murder_mysteries | 1|none | 0|acc_norm |↑ |0.5160|± |0.0317|
|
|
| - leaderboard_musr_object_placements | 1|none | 0|acc_norm |↑ |0.2695|± |0.0278|
|
|
| - leaderboard_musr_team_allocation | 1|none | 0|acc_norm |↑ |0.3480|± |0.0302|
|
|
|
|
```bash
|
|
litgpt evaluate --tasks 'gsm8k,mathqa' --out_dir 'evaluate-math/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
|
|
```
|
|
|
|
|Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr|
|
|
|------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|
|
|gsm8k | 3|flexible-extract| 5|exact_match|↑ |0.0205|± |0.0039|
|
|
| | |strict-match | 5|exact_match|↑ |0.0000|± |0.0000|
|
|
|mathqa| 1|none | 0|acc |↑ |0.2010|± |0.0073|
|
|
| | |none | 0|acc_norm |↑ |0.2077|± |0.0074|
|
|
|
|
```bash
|
|
litgpt evaluate --tasks 'mmlu,mmlu_pro' --out_dir 'evaluate-mmlu/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
|
|
```
|
|
|
|
| Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr|
|
|
|---------------------------------------|------:|--------------|-----:|-----------|---|-----:|---|-----:|
|
|
|mmlu | 2|none | |acc |↑ |0.2468|± |0.0036|
|
|
| - humanities | 2|none | |acc |↑ |0.2459|± |0.0063|
|
|
| - formal_logic | 1|none | 0|acc |↑ |0.3175|± |0.0416|
|
|
| - high_school_european_history | 1|none | 0|acc |↑ |0.2364|± |0.0332|
|
|
| - high_school_us_history | 1|none | 0|acc |↑ |0.2304|± |0.0296|
|
|
| - high_school_world_history | 1|none | 0|acc |↑ |0.2194|± |0.0269|
|
|
| - international_law | 1|none | 0|acc |↑ |0.2479|± |0.0394|
|
|
| - jurisprudence | 1|none | 0|acc |↑ |0.2315|± |0.0408|
|
|
| - logical_fallacies | 1|none | 0|acc |↑ |0.2147|± |0.0323|
|
|
| - moral_disputes | 1|none | 0|acc |↑ |0.2168|± |0.0222|
|
|
| - moral_scenarios | 1|none | 0|acc |↑ |0.2726|± |0.0149|
|
|
| - philosophy | 1|none | 0|acc |↑ |0.1865|± |0.0221|
|
|
| - prehistory | 1|none | 0|acc |↑ |0.2191|± |0.0230|
|
|
| - professional_law | 1|none | 0|acc |↑ |0.2490|± |0.0110|
|
|
| - world_religions | 1|none | 0|acc |↑ |0.3450|± |0.0365|
|
|
| - other | 2|none | |acc |↑ |0.2385|± |0.0076|
|
|
| - business_ethics | 1|none | 0|acc |↑ |0.2200|± |0.0416|
|
|
| - clinical_knowledge | 1|none | 0|acc |↑ |0.2264|± |0.0258|
|
|
| - college_medicine | 1|none | 0|acc |↑ |0.2601|± |0.0335|
|
|
| - global_facts | 1|none | 0|acc |↑ |0.1900|± |0.0394|
|
|
| - human_aging | 1|none | 0|acc |↑ |0.2422|± |0.0288|
|
|
| - management | 1|none | 0|acc |↑ |0.2330|± |0.0419|
|
|
| - marketing | 1|none | 0|acc |↑ |0.2821|± |0.0295|
|
|
| - medical_genetics | 1|none | 0|acc |↑ |0.2900|± |0.0456|
|
|
| - miscellaneous | 1|none | 0|acc |↑ |0.2388|± |0.0152|
|
|
| - nutrition | 1|none | 0|acc |↑ |0.1993|± |0.0229|
|
|
| - professional_accounting | 1|none | 0|acc |↑ |0.2270|± |0.0250|
|
|
| - professional_medicine | 1|none | 0|acc |↑ |0.2610|± |0.0267|
|
|
| - virology | 1|none | 0|acc |↑ |0.2349|± |0.0330|
|
|
| - social sciences | 2|none | |acc |↑ |0.2632|± |0.0079|
|
|
| - econometrics | 1|none | 0|acc |↑ |0.2544|± |0.0410|
|
|
| - high_school_geography | 1|none | 0|acc |↑ |0.1869|± |0.0278|
|
|
| - high_school_government_and_politics| 1|none | 0|acc |↑ |0.2850|± |0.0326|
|
|
| - high_school_macroeconomics | 1|none | 0|acc |↑ |0.3128|± |0.0235|
|
|
| - high_school_microeconomics | 1|none | 0|acc |↑ |0.2773|± |0.0291|
|
|
| - high_school_psychology | 1|none | 0|acc |↑ |0.2422|± |0.0184|
|
|
| - human_sexuality | 1|none | 0|acc |↑ |0.2595|± |0.0384|
|
|
| - professional_psychology | 1|none | 0|acc |↑ |0.2435|± |0.0174|
|
|
| - public_relations | 1|none | 0|acc |↑ |0.2273|± |0.0401|
|
|
| - security_studies | 1|none | 0|acc |↑ |0.3265|± |0.0300|
|
|
| - sociology | 1|none | 0|acc |↑ |0.2537|± |0.0308|
|
|
| - us_foreign_policy | 1|none | 0|acc |↑ |0.3000|± |0.0461|
|
|
| - stem | 2|none | |acc |↑ |0.2404|± |0.0076|
|
|
| - abstract_algebra | 1|none | 0|acc |↑ |0.1700|± |0.0378|
|
|
| - anatomy | 1|none | 0|acc |↑ |0.2074|± |0.0350|
|
|
| - astronomy | 1|none | 0|acc |↑ |0.2105|± |0.0332|
|
|
| - college_biology | 1|none | 0|acc |↑ |0.2153|± |0.0344|
|
|
| - college_chemistry | 1|none | 0|acc |↑ |0.2000|± |0.0402|
|
|
| - college_computer_science | 1|none | 0|acc |↑ |0.2300|± |0.0423|
|
|
| - college_mathematics | 1|none | 0|acc |↑ |0.1700|± |0.0378|
|
|
| - college_physics | 1|none | 0|acc |↑ |0.2647|± |0.0439|
|
|
| - computer_security | 1|none | 0|acc |↑ |0.2700|± |0.0446|
|
|
| - conceptual_physics | 1|none | 0|acc |↑ |0.2766|± |0.0292|
|
|
| - electrical_engineering | 1|none | 0|acc |↑ |0.2552|± |0.0363|
|
|
| - elementary_mathematics | 1|none | 0|acc |↑ |0.2566|± |0.0225|
|
|
| - high_school_biology | 1|none | 0|acc |↑ |0.2097|± |0.0232|
|
|
| - high_school_chemistry | 1|none | 0|acc |↑ |0.2611|± |0.0309|
|
|
| - high_school_computer_science | 1|none | 0|acc |↑ |0.2600|± |0.0441|
|
|
| - high_school_mathematics | 1|none | 0|acc |↑ |0.2111|± |0.0249|
|
|
| - high_school_physics | 1|none | 0|acc |↑ |0.2517|± |0.0354|
|
|
| - high_school_statistics | 1|none | 0|acc |↑ |0.3056|± |0.0314|
|
|
| - machine_learning | 1|none | 0|acc |↑ |0.2857|± |0.0429|
|
|
|mmlu_pro | 2|custom-extract| |exact_match|↑ |0.0000|± |0.0000|
|
|
| - biology | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - business | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - chemistry | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - computer_science | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - economics | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - engineering | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - health | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - history | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - law | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - math | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - other | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - philosophy | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - physics | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
| - psychology | 1|custom-extract| 5|exact_match|↑ |0.0000|± |0.0000|
|
|
|
|
| Groups |Version| Filter |n-shot| Metric | |Value | |Stderr|
|
|
|------------------|------:|--------------|------|-----------|---|-----:|---|-----:|
|
|
|mmlu | 2|none | |acc |↑ |0.2468|± |0.0036|
|
|
| - humanities | 2|none | |acc |↑ |0.2459|± |0.0063|
|
|
| - other | 2|none | |acc |↑ |0.2385|± |0.0076|
|
|
| - social sciences| 2|none | |acc |↑ |0.2632|± |0.0079|
|
|
| - stem | 2|none | |acc |↑ |0.2404|± |0.0076|
|
|
|mmlu_pro | 2|custom-extract| |exact_match|↑ |0.0000|± |0.0000|
|
|
|
|
```bash
|
|
litgpt evaluate --tasks 'arc_challenge,boolq,gpqa,hellaswag,openbookqa,piqa,truthfulqa_mc2,winogrande' --out_dir 'evaluate-reasoning/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
|
|
```
|
|
|
|
| Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr|
|
|
|-------------------------------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|
|
|arc_challenge | 1|none | 0|acc |↑ |0.1852|± |0.0114|
|
|
| | |none | 0|acc_norm |↑ |0.2201|± |0.0121|
|
|
|boolq | 2|none | 0|acc |↑ |0.4446|± |0.0087|
|
|
|gpqa_diamond_cot_n_shot | 2|flexible-extract| 0|exact_match|↑ |0.0859|± |0.0200|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_diamond_cot_zeroshot | 1|flexible-extract| 0|exact_match|↑ |0.0606|± |0.0170|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_diamond_generative_n_shot | 2|flexible-extract| 0|exact_match|↑ |0.1717|± |0.0269|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_diamond_n_shot | 2|none | 0|acc |↑ |0.2677|± |0.0315|
|
|
| | |none | 0|acc_norm |↑ |0.2677|± |0.0315|
|
|
|gpqa_diamond_zeroshot | 1|none | 0|acc |↑ |0.1970|± |0.0283|
|
|
| | |none | 0|acc_norm |↑ |0.1970|± |0.0283|
|
|
|gpqa_extended_cot_n_shot | 2|flexible-extract| 0|exact_match|↑ |0.0971|± |0.0127|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_extended_cot_zeroshot | 1|flexible-extract| 0|exact_match|↑ |0.0696|± |0.0109|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_extended_generative_n_shot| 2|flexible-extract| 0|exact_match|↑ |0.1502|± |0.0153|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_extended_n_shot | 2|none | 0|acc |↑ |0.2399|± |0.0183|
|
|
| | |none | 0|acc_norm |↑ |0.2399|± |0.0183|
|
|
|gpqa_extended_zeroshot | 1|none | 0|acc |↑ |0.2473|± |0.0185|
|
|
| | |none | 0|acc_norm |↑ |0.2473|± |0.0185|
|
|
|gpqa_main_cot_n_shot | 2|flexible-extract| 0|exact_match|↑ |0.1116|± |0.0149|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_main_cot_zeroshot | 1|flexible-extract| 0|exact_match|↑ |0.0625|± |0.0114|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_main_generative_n_shot | 2|flexible-extract| 0|exact_match|↑ |0.1384|± |0.0163|
|
|
| | |strict-match | 0|exact_match|↑ |0.0000|± |0.0000|
|
|
|gpqa_main_n_shot | 2|none | 0|acc |↑ |0.2388|± |0.0202|
|
|
| | |none | 0|acc_norm |↑ |0.2388|± |0.0202|
|
|
|gpqa_main_zeroshot | 1|none | 0|acc |↑ |0.2500|± |0.0205|
|
|
| | |none | 0|acc_norm |↑ |0.2500|± |0.0205|
|
|
|hellaswag | 1|none | 0|acc |↑ |0.2628|± |0.0044|
|
|
| | |none | 0|acc_norm |↑ |0.2705|± |0.0044|
|
|
|openbookqa | 1|none | 0|acc |↑ |0.1360|± |0.0153|
|
|
| | |none | 0|acc_norm |↑ |0.2620|± |0.0197|
|
|
|piqa | 1|none | 0|acc |↑ |0.5550|± |0.0116|
|
|
| | |none | 0|acc_norm |↑ |0.5528|± |0.0116|
|
|
|truthfulqa_mc2 | 2|none | 0|acc |↑ |0.5010|± |0.0159|
|
|
|winogrande | 1|none | 0|acc |↑ |0.5130|± |0.0140|
|
|
|
|
```bash
|
|
litgpt evaluate --tasks 'wikitext,qasper' --out_dir 'evaluate-long/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
|
|
```
|
|
|
|
| Tasks |Version|Filter|n-shot| Metric | | Value | |Stderr|
|
|
|---------------|------:|------|-----:|---------------|---|--------:|---|------|
|
|
|qasper_bool | 1|none | 0|f1 |↑ | 0.8966|± |0.0166|
|
|
|qasper_freeform| 2|none | 0|f1_abstractive |↑ | 0.0597|± |0.0052|
|
|
|wikitext | 2|none | 0|bits_per_byte |↓ | 2.2154|± | N/A|
|
|
| | |none | 0|byte_perplexity|↓ | 4.6441|± | N/A|
|
|
| | |none | 0|word_perplexity|↓ |3683.1019|± | N/A|
|
|
|
