mtasic85's picture
pretrain model
81565a0
|
raw
history blame
3.47 kB
metadata
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

logo

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

val_ppl

epoch

learning_rate

lm-evaluation-harness

litgpt evaluate --tasks 'hellaswag,gsm8k,truthfulqa_mc2,mmlu,winogrande,arc_challenge' --out_dir 'evaluate-quick/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
litgpt evaluate --tasks 'leaderboard' --out_dir 'evaluate-leaderboard/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
litgpt evaluate --tasks 'bbh_zeroshot,bbh_fewshot,bbh_cot_fewshot,bbh_cot_zeroshot' --out_dir 'evaluate-bigbenchhard/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
litgpt evaluate --tasks 'mmlu,mmlu_pro' --out_dir 'evaluate-mmlu/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
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/
litgpt evaluate --tasks 'mmlu_multilingual,mgsm' --out_dir 'evaluate-multilinguals/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
litgpt evaluate --tasks 'gsm8k,mathqa' --out_dir 'evaluate-math/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/
litgpt evaluate --tasks 'wikitext,qasper' --out_dir 'evaluate-long/' --batch_size 4 --dtype 'bfloat16' out/pretrain/final/