mirror of
https://github.com/QwenLM/Qwen.git
synced 2026-05-20 16:35:47 +08:00
add gradient checkpointing
This commit is contained in:
@@ -44,4 +44,5 @@ torchrun $DISTRIBUTED_ARGS finetune.py \
|
||||
--model_max_length 2048 \
|
||||
--lazy_preprocess True \
|
||||
--use_lora \
|
||||
--gradient_checkpointing \
|
||||
--deepspeed finetune/ds_config_zero2.json
|
||||
Reference in New Issue
Block a user