InternLM/internlm
YWMditto 28635755f5
[fix bug] Fix the error that RotaryEmbedding is converted to a non-fp32 format during training, and add a compatible method for the llama model. (#239)
Co-authored-by: YWMditto <862779238@qq.com>
2023-08-26 17:48:08 +08:00
..
apis initial commit 2023-07-06 12:55:23 +08:00
core Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
data Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
initialize Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
model [fix bug] Fix the error that RotaryEmbedding is converted to a non-fp32 format during training, and add a compatible method for the llama model. (#239) 2023-08-26 17:48:08 +08:00
monitor Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
solver Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
train Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
utils Merge develop to main (#233) 2023-08-24 22:03:04 +08:00
__init__.py initial commit 2023-07-06 12:55:23 +08:00