mirror of
https://github.com/hiyouga/LLaMA-Factory.git
synced 2025-12-23 23:30:36 +08:00
47 lines
1.0 KiB
YAML
47 lines
1.0 KiB
YAML
# Start FSDP fine-tuning
|
|
# accelerate launch \
|
|
# --config_file examples/accelerate/fsdp_config.yaml \
|
|
# src/train.py examples/ascend/qwen3moe_full_sft_fsdp.yaml
|
|
# Change `num_processes` in fsdp_config.yaml to 16 in A3
|
|
|
|
### model
|
|
model_name_or_path: Qwen/Qwen3-30B-A3B-Instruct-2507
|
|
trust_remote_code: true
|
|
use_v1_kernels: true
|
|
flash_attn: fa2
|
|
|
|
### method
|
|
stage: sft
|
|
do_train: true
|
|
finetuning_type: full
|
|
disable_gradient_checkpointing: false
|
|
|
|
### dataset
|
|
dataset: alpaca_zh
|
|
template: qwen3
|
|
cutoff_len: 1024
|
|
overwrite_cache: true
|
|
preprocessing_num_workers: 16
|
|
dataloader_num_workers: 4
|
|
|
|
### output
|
|
output_dir: saves/Qwen3-30B-A3B-Instruct-2507/full/sft
|
|
logging_steps: 1
|
|
save_steps: 500
|
|
max_steps: 500
|
|
plot_loss: true
|
|
overwrite_output_dir: true
|
|
save_only_model: true
|
|
report_to: none # choices: [none, wandb, tensorboard, swanlab, mlflow]
|
|
|
|
### train
|
|
per_device_train_batch_size: 4
|
|
gradient_accumulation_steps: 1
|
|
learning_rate: 1.0e-4
|
|
lr_scheduler_type: cosine
|
|
warmup_ratio: 0.1
|
|
bf16: true
|
|
ddp_timeout: 180000000
|
|
resume_from_checkpoint: null
|
|
seed: 1234
|