mirror of
https://github.com/hiyouga/LLaMA-Factory.git
synced 2026-07-28 11:46:09 +08:00
b61140db3e
Co-authored-by: frozenleaves <frozenleaves@users.noreply.github.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
24 lines
368 B
YAML
24 lines
368 B
YAML
model: Qwen/Qwen3-0.6B
|
|
trust_remote_code: true
|
|
model_class: llm
|
|
|
|
flash_attn: flash_attention_2
|
|
|
|
# FSDP Config
|
|
dist_config:
|
|
name: fsdp2
|
|
dcp_path: null
|
|
cp_mode: ulysses
|
|
cp_size: 2
|
|
|
|
### data
|
|
train_dataset: data/v1_sft_demo.yaml
|
|
|
|
### training
|
|
output_dir: outputs/test_ulysses_cp
|
|
micro_batch_size: 1
|
|
cutoff_len: 2048
|
|
learning_rate: 1.0e-4
|
|
bf16: false
|
|
max_steps: 10
|