Sft-new ssurface/qwen3-4b-sft-l5-merged 4B • Updated 3 days ago • 19 ssurface/qwen3-4b-sft-l3-merged 4B • Updated 3 days ago • 18 ssurface/qwen3-4b-sft-l4-merged 4B • Updated 3 days ago • 16 ssurface/qwen3-4b-sft-l2-merged 4B • Updated 3 days ago • 16
GRPO no length punshiment SFT GSM8K ssurface/qwen3-4b-grpo-nolength-l1 Text Generation • 4B • Updated 9 days ago • 13 ssurface/qwen3-4b-grpo-nolength-l2 Text Generation • 4B • Updated 9 days ago • 19 ssurface/qwen3-4b-grpo-nolength-l3 Text Generation • 4B • Updated 9 days ago • 20 ssurface/qwen3-4b-grpo-nolength-l4 Text Generation • 4B • Updated 9 days ago • 18
GRPO SFT GSM8K ssurface/qwen3-4b-grpo-l1 Text Generation • 4B • Updated 15 days ago • 18 ssurface/qwen3-4b-grpo-l2 Text Generation • 4B • Updated 15 days ago • 18 ssurface/qwen3-4b-grpo-l3 Text Generation • 4B • Updated 15 days ago • 23 ssurface/qwen3-4b-grpo-l4 Text Generation • 4B • Updated 15 days ago • 16
GRPO SFT-Length-Punishment-GDPO SFT GSM8K ssurface/qwen3-4b-gdpo-length-sft-l1 Text Generation • 4B • Updated 8 days ago • 251 ssurface/qwen3-4b-gdpo-length-sft-l2 Text Generation • 4B • Updated 8 days ago • 252 ssurface/qwen3-4b-gdpo-length-sft-l3 Text Generation • 4B • Updated 8 days ago • 252 ssurface/qwen3-4b-gdpo-length-sft-l4 Text Generation • 4B • Updated 8 days ago • 260
GRPO abstract reward SFT GSM8K ssurface/qwen3-4b-grpo-abstract-l1 Text Generation • 4B • Updated 12 days ago • 24 ssurface/qwen3-4b-grpo-abstract-l2 Text Generation • 4B • Updated 12 days ago • 23 ssurface/qwen3-4b-grpo-abstract-l3 Text Generation • 4B • Updated 12 days ago • 24 ssurface/qwen3-4b-grpo-abstract-l4 Text Generation • 4B • Updated 12 days ago • 24
Qwen3-4B CoT Compression Study LoRA adapters trained for 5 progressively shorter chain-of-thought styles on GSM8K, plus the eval artifacts behind the Pareto curve. ssurface/qwen3-4b-cot-compress-l1 Text Generation • Updated May 25 • 17 ssurface/qwen3-4b-cot-compress-l2 Text Generation • Updated May 25 • 5 ssurface/qwen3-4b-cot-compress-l3 Text Generation • Updated May 25 • 5 ssurface/qwen3-4b-cot-compress-l4 Text Generation • Updated May 25 • 4
Sft-new ssurface/qwen3-4b-sft-l5-merged 4B • Updated 3 days ago • 19 ssurface/qwen3-4b-sft-l3-merged 4B • Updated 3 days ago • 18 ssurface/qwen3-4b-sft-l4-merged 4B • Updated 3 days ago • 16 ssurface/qwen3-4b-sft-l2-merged 4B • Updated 3 days ago • 16
GRPO SFT-Length-Punishment-GDPO SFT GSM8K ssurface/qwen3-4b-gdpo-length-sft-l1 Text Generation • 4B • Updated 8 days ago • 251 ssurface/qwen3-4b-gdpo-length-sft-l2 Text Generation • 4B • Updated 8 days ago • 252 ssurface/qwen3-4b-gdpo-length-sft-l3 Text Generation • 4B • Updated 8 days ago • 252 ssurface/qwen3-4b-gdpo-length-sft-l4 Text Generation • 4B • Updated 8 days ago • 260
GRPO no length punshiment SFT GSM8K ssurface/qwen3-4b-grpo-nolength-l1 Text Generation • 4B • Updated 9 days ago • 13 ssurface/qwen3-4b-grpo-nolength-l2 Text Generation • 4B • Updated 9 days ago • 19 ssurface/qwen3-4b-grpo-nolength-l3 Text Generation • 4B • Updated 9 days ago • 20 ssurface/qwen3-4b-grpo-nolength-l4 Text Generation • 4B • Updated 9 days ago • 18
GRPO abstract reward SFT GSM8K ssurface/qwen3-4b-grpo-abstract-l1 Text Generation • 4B • Updated 12 days ago • 24 ssurface/qwen3-4b-grpo-abstract-l2 Text Generation • 4B • Updated 12 days ago • 23 ssurface/qwen3-4b-grpo-abstract-l3 Text Generation • 4B • Updated 12 days ago • 24 ssurface/qwen3-4b-grpo-abstract-l4 Text Generation • 4B • Updated 12 days ago • 24
GRPO SFT GSM8K ssurface/qwen3-4b-grpo-l1 Text Generation • 4B • Updated 15 days ago • 18 ssurface/qwen3-4b-grpo-l2 Text Generation • 4B • Updated 15 days ago • 18 ssurface/qwen3-4b-grpo-l3 Text Generation • 4B • Updated 15 days ago • 23 ssurface/qwen3-4b-grpo-l4 Text Generation • 4B • Updated 15 days ago • 16
Qwen3-4B CoT Compression Study LoRA adapters trained for 5 progressively shorter chain-of-thought styles on GSM8K, plus the eval artifacts behind the Pareto curve. ssurface/qwen3-4b-cot-compress-l1 Text Generation • Updated May 25 • 17 ssurface/qwen3-4b-cot-compress-l2 Text Generation • Updated May 25 • 5 ssurface/qwen3-4b-cot-compress-l3 Text Generation • Updated May 25 • 5 ssurface/qwen3-4b-cot-compress-l4 Text Generation • Updated May 25 • 4
