marx/llama-recipes @ 1437b917d0322465a24d4aa1126cd58199848851

Matthias Reso 9def4fbafd Remove micro_batch_training parameter and replace with gradient_accumulation_steps		1 gadu atpakaļ
..
__init__.py	4767f09ecd Initial commit	1 gadu atpakaļ
datasets.py	4767f09ecd Initial commit	1 gadu atpakaļ
fsdp.py	a977145a9b change bf16 default to false	1 gadu atpakaļ
peft.py	4767f09ecd Initial commit	1 gadu atpakaļ
training.py	9def4fbafd Remove micro_batch_training parameter and replace with gradient_accumulation_steps	1 gadu atpakaļ