Pinned Repositories
dsir_japanese
DSIR large-scale data selection framework for language model training (Japanese tokenizer)
japanese_llm_eval
A repo for evaluating Japanese LLMs ・ 日本語LLMを評価するレポ
lm-evaluation-harness
A framework for few-shot evaluation of autoregressive language models.
lm-evaluation-harness-eleuther
A framework for few-shot evaluation of language models.
mao-seminar
書籍「わかばちゃんと学ぶGit使い方入門」内で使う、ゼミのWebサイトのソースコードです。プルリクエストの練習は「pull-request-practiceリポジトリ」で行ってください。URL: https://github.com/elmas3/pull-request-practice
multilingual-mt-bench
qlora
QLoRA: Efficient Finetuning of Quantized LLMs
shitsu
A fast multilingual text quality classifier
suzume
trl
Train transformer language models with reinforcement learning.
lightblue-tech's Repositories
lightblue-tech/shitsu
A fast multilingual text quality classifier
lightblue-tech/japanese_llm_eval
A repo for evaluating Japanese LLMs ・ 日本語LLMを評価するレポ
lightblue-tech/suzume
lightblue-tech/dsir_japanese
DSIR large-scale data selection framework for language model training (Japanese tokenizer)
lightblue-tech/lm-evaluation-harness
A framework for few-shot evaluation of autoregressive language models.
lightblue-tech/lm-evaluation-harness-eleuther
A framework for few-shot evaluation of language models.
lightblue-tech/mao-seminar
書籍「わかばちゃんと学ぶGit使い方入門」内で使う、ゼミのWebサイトのソースコードです。プルリクエストの練習は「pull-request-practiceリポジトリ」で行ってください。URL: https://github.com/elmas3/pull-request-practice
lightblue-tech/multilingual-mt-bench
lightblue-tech/qlora
QLoRA: Efficient Finetuning of Quantized LLMs
lightblue-tech/trl
Train transformer language models with reinforcement learning.