423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-03-01 17:26:03 +08:00
Code Issues Packages Projects Releases Wiki Activity

[model] update constants (#10220)

Browse Source
This commit is contained in:
Yaowei Zheng
2026-02-26 21:13:56 +08:00
committed by GitHub
parent 2b8b871475
commit 122cd46084
12 changed files with 69 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/v1/train_full/train_full_deepspeed.yaml
View File

@@ -6,14 +6,14 @@ template: qwen3_nothink
kernel_config:
name: auto
include_kernels: auto
include_kernels: auto
dist_config:
name: deepspeed
config_file: examples/deepspeed/ds_z3_config.json
### data
train_dataset: data/v1_sft_demo.yaml
train_dataset: data/v1_sft_demo.yaml
### training
output_dir: outputs/Qwen3-0.6B-deepspeed
@@ -22,4 +22,3 @@ cutoff_len: 2048
learning_rate: 1.0e-4
bf16: true
max_steps: 10