193 GB

Ctrl+K

1 contributor

History: 23 commits

xingxm

Add eval artifacts: summary, per-case scores, significance tests, rubric stats, HTML report

eb98c6f verified about 2 months ago

designcoder_qwen3.5_4b_adamw_bs256_data41287_step200
Add designcoder_qwen3.5_4b_adamw_bs256_data41287_step200 (bench-200: 84.22) about 2 months ago
designcoder_qwen3.5_4b_muon_bs256_data41287_step200
Add designcoder_qwen3.5_4b_muon_bs256_data41287_step200 (bench-200: 83.36) about 2 months ago
designcoder_qwen3.5_4b_muon_bs32_step1900
Add designcoder_qwen3.5_4b_muon_bs32_step1900 (base Qwen3.5-4B, Muon, lr1e-5, global bs32, 2 epochs, step 1900) 2 months ago
designcoder_qwen3.5_9b_adamw_bs16_step3800
Add designcoder_qwen3.5_9b_adamw_bs16_step3800 (base Qwen3.5-9B, AdamW, lr2e-5, global bs16, 2 epochs, step 3800) 2 months ago
designcoder_qwen3.5_9b_adamw_bs256_data41287_step200
Add designcoder_qwen3.5_9b_adamw_bs256_data41287_step200 (bench-200: 84.40) about 2 months ago
designcoder_qwen3.5_9b_muon_bs16_step3800
Add training log for 9b muon run 2 months ago
designcoder_qwen3.6_27b_adamw_bs32_step1900
Add designcoder_qwen3.6_27b_adamw_bs32_step1900 (base Qwen3.6-27B, AdamW, lr1e-5, global bs32, 2 epochs, step 1900) 2 months ago
designcoder_qwen3.8_27b_adamw_bs128_data41287_step400
Add designcoder_qwen3.8_27b_adamw_bs128_data41287_step400 (bench-200: 91.19, strongest in release) about 2 months ago
eval
Add eval artifacts: summary, per-case scores, significance tests, rubric stats, HTML report about 2 months ago
.gitattributes

2.51 kB
Add designcoder_qwen3.8_27b_adamw_bs128_data41287_step400 (bench-200: 91.19, strongest in release) about 2 months ago
README.md

8.57 kB
Update bench-200 final full-run scores (n=200 all models) + eval artifacts about 2 months ago