/

/

karpathy_nanoGPT

Обзор ДокументацияВойти

/

/

karpathy_nanoGPT

Аналитика

Безопасность

karpathy_nanoGPT

Публичный

Описание

The simplest, fastest repository for training/finetuning medium-sized GPTs.

91.9 MiB

Лицензия

В избранном0

Языки

Python100%

Сообщить о нарушении

Коммиты:

Andrej

Merge pull request #420 from vinjn/fix-371-enc-is-not-defined

27 фев 2024, 20:27

Не верифицирован

27 фев 2024, 20:27325be85

adjust teaser figure with a more tuned result

4 года назад

Fix for gradient_accumulation_steps training slow

3 года назад

Merge pull request #420 from vinjn/fix-371-enc-is-not-defined

3 года назад

keep only what's needed

4 года назад

feature: .gitignore - added venv folders

3 года назад

Add MIT LICENSE file

4 года назад

docs: simplify dependencies installation

3 года назад

Fix AssertionError on macOS - need to check CUDA availability for bf16

3 года назад

configurator.py

shuttling the poor mans configurator aside into its own file and adding it to all of train,sample,bench. because i am leaving args in globals() so i can avoid having to prepend every single variable with an args., i have to exec the configurator and the optional configs. so we're left with something very gross by standard convention but also quite simple and functional. *ducks*

4 года назад

Merge pull request #274 from apivovarov/gelu

3 года назад

Fix AssertionError on macOS - need to check CUDA availability for bf16

3 года назад

scaling_laws.ipynb

fix typo ( params -> tokens)

4 года назад

fix np.memmap memory leak

3 года назад

transformer_sizing.ipynb

oops forgot to subtract embedding params, which don't enter the 6ND equation

4 года назад

README.md