Spaces:

Dovakiins
/

qwerrwe

Build error

App Files Files Community

qwerrwe / src /axolotl /utils

100 contributors

History: 542 commits

Keith Stevens

Generalizing the chat_template prompt strategy (#1660) [skip ci]

cc11c6b unverified 10 months ago

callbacks
fixes to save on fractional save_steps (#1643) 10 months ago
config
Generalizing the chat_template prompt strategy (#1660) [skip ci] 10 months ago
data
Switch to parallel FFD bin packing algorithm. (#1619) 10 months ago
gradient_checkpointing
Unsloth gradient checkpointing offload (#1528) 11 months ago
samplers
Switch to parallel FFD bin packing algorithm. (#1619) 10 months ago
__init__.py

136 Bytes

WIP: Support table logging for mlflow, too (#1506) 12 months ago
bench.py

2.44 kB

FDSP + QLoRA (#1378) about 1 year ago
chat_templates.py

4.13 kB

Fix llama3 chat_template (extra <|eot_id|> on last turn) (#1635) 10 months ago
collators.py

9.73 kB

fix attention mask collation (#1603) 11 months ago
dict.py

327 Bytes

Pydantic 2.x cfg (#1239) about 1 year ago
distributed.py

7.14 kB

DBRX Model Support (#1462) 12 months ago
freeze.py

8.33 kB

ORPO (#1419) about 1 year ago
lora_embeddings.py

375 Bytes

Phi2 multipack (#1173) about 1 year ago
mlflow_.py

543 Bytes

ADD: push checkpoints to mlflow artifact registry (#1295) [skip ci] about 1 year ago
model_shard_quant.py

8.95 kB

DBRX Model Support (#1462) 12 months ago
models.py

38.4 kB

Add KTO support (#1640) 10 months ago
schedulers.py

7.66 kB

Scheduler implementation of Continual Pre-Training of Large Language Models: How to (re)warm your model? (#1273) about 1 year ago
tokenization.py

5.16 kB

Add debug option for RL dataset preprocessing (#1404) 11 months ago
trainer.py

16.2 kB

Switch to parallel FFD bin packing algorithm. (#1619) 10 months ago
wandb_.py

609 Bytes

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