GitBross Página inicial Explorar Ajuda
Registrar Entrar
sol_87yf6cwp
/
LLMs-from-scratch
mirror de https://github.com/br8bit/LLMs-from-scratch.git
1
0
Fork 0
Arquivos Issues 0 Wiki
Branch: main
Branches Tags
LLMs-from-scrat...
/
ch05
/
05_bonus_hparam_tuning
Sebastian Raschka 8b3e4b24b0 Remove unused params for hparam script (#710) 4 meses atrás
..
README.md e757091301 Organized setup instructions (#115) 1 ano atrás
hparam_search.py 8b3e4b24b0 Remove unused params for hparam script (#710) 4 meses atrás
the-verdict.txt cf39abac04 Add and link bonus material (#84) 1 ano atrás

README.md

Optimizing Hyperparameters for Pretraining

The hparam_search.py script, based on the extended training function in Appendix D: Adding Bells and Whistles to the Training Loop, is designed to find optimal hyperparameters via grid search.

[!NOTE] This script will take a long time to run. You may want to reduce the number of hyperparameter configurations explored in the HPARAM_GRID dictionary at the top.