Spaces:

Dovakiins
/

qwerrwe

Build error

App Files Files Community

qwerrwe / src /axolotl /utils

Ctrl+K

Ctrl+K

100 contributors

History: 425 commits

winglian's picture

improve vram use w gradient checkpointing (#1167) [skip ci]

802f966 unverified over 1 year ago

samplers
Efficiently get the length of the tokenized docs (#1063) over 1 year ago
__init__.py

0 Bytes

WIP large refactor to make finetune script a little more manageable (#3) about 2 years ago
bench.py
2.09 kB

jupyter lab fixes (#1139) [skip ci] over 1 year ago
callbacks.py
23.3 kB

Add mlflow callback for pushing config to mlflow artifacts (#1125) over 1 year ago
chat_templates.py
1.52 kB

Adds chat templates (#1022) over 1 year ago
collators.py

8.96 kB

Multipack simplify for Mixtral (#1142) over 1 year ago
config.py
20.6 kB

improve vram use w gradient checkpointing (#1167) [skip ci] over 1 year ago
data.py
29.8 kB

Deprecate max packed sequence len (#1141) over 1 year ago
dict.py

326 Bytes

fix DefaultDict.__or__ almost 2 years ago
distributed.py

7.07 kB

cleanup verbosity a bit over 1 year ago
freeze.py

1.25 kB

Fix Deepspeed loading (#950) over 1 year ago
lora_embeddings.py

377 Bytes

add gptneox embeddings, fix phi2 inputs, also fix the casting (#1083) over 1 year ago
mlflow_.py

511 Bytes

Add: mlflow for experiment tracking (#1059) [skip ci] over 1 year ago
models.py
31 kB

jupyter lab fixes (#1139) [skip ci] over 1 year ago
schedulers.py

4.75 kB

Cosine learning rate schedule - minimum learning rate (#1062) over 1 year ago
tokenization.py

1.43 kB

add a latest tag for regular axolotl image, cleanup extraneous print statement (#746) over 1 year ago
trainer.py

11.2 kB

Deprecate max packed sequence len (#1141) over 1 year ago
wandb_.py

609 Bytes

Feat(wandb): Refactor to be more flexible (#767) over 1 year ago