JackHsieh/sft_with_kl_qwen-4B_NR-short-32k-16-1k-8_lr-3e-06-constant-bs-512_steps-128
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: JackHsieh/sft_with_kl_qwen-4B_NR-short-32k-16-1k-8_lr-3e-06-constant-bs-512_steps-128
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-20 17:57:21 +08:00
commit 611af498ba
14 changed files with 152247 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
README.md Normal file
View File

@@ -0,0 +1,2 @@
Checkpoint saved from [sft-with-kl/qwen-4B/NR-short-with-offline-thoughts/32k-16-thoughts-1k-8-thoughts/sft/lr=3e-06-constant-bs=512-epochs=5-thought_coef=0.01](https://wandb.ai/jackellishsieh-stanford-university/prestar/groups/sft-with-kl/runs/jgi1xm3h?nw=nwuserjackellishsieh), step 128.