jaeh8nkim/s1K1p1-Distillepoch15-Qwen3-0.6B
0.6B
•
Updated
•
1
Note
Qwen3-0.6B + s1K-1.1
jaeh8nkim/s1Kteacher8KUP-Distill-Qwen3-0.6B
0.6B
•
Updated
•
2
Note
Qwen3-0.6B + Teacher-generated
jaeh8nkim/s1Kstudent8KUP-Distill-Qwen3-0.6B
0.6B
•
Updated
•
3
Note
Qwen3-0.6B + Self-distill
jaeh8nkim/s1Kskd8KUP-Distill-Qwen3-0.6B
0.6B
•
Updated
•
2
Note
Qwen3-0.6B + SKD-inspired
jaeh8nkim/s1K4Q3p6BUPFTstep1prob10-Distillepoch15-Qwen3-0.6B
0.6B
•
Updated
•
2
Note
Qwen3-0.6B + RSD-generated (p_th=10%)
jaeh8nkim/s4Qp6UPst1pr3ep15-Distill-Qwen3-0.6B
0.6B
•
Updated
•
1
Note
Qwen3-0.6B + RSD-generated (p_th=3%)
jaeh8nkim/s1K4Q3p6Bs1p17BtUPFTstep1epoch15-Distill-Qwen3-0.6B
0.6B
•
Updated
•
1
Note
Qwen3-0.6B + RSD-generated (p_th=1%)
jaeh8nkim/s4Qp6UPst1pr03ep15-Distill-Qwen3-0.6B
0.6B
•
Updated
•
1
Note
Qwen3-0.6B + RSD-generated (p_th=0.3%)
jaeh8nkim/s1K4Q317UP-Distill-Qwen3-1.7B
2B
•
Updated
•
2
Note
Qwen3-1.7B + RSD-generated (p_th=1%) tailored for Qwen3-1.7B
jaeh8nkim/s1K4L321UP-Distill-Llama-3.2-1B-Instruct
1B
•
Updated
•
2
Note
Llama-3.2-1B-Instruct + RSD-generated (p_th=1%) tailored for Llama-3.2-1B-Instruct
jaeh8nkim/s1Kstudent203UP-Distill-Qwen3-0.6B
0.6B
•
Updated
•
6
Note
Qwen3-0.6B + Self-distill (203 rejection sampling attempts)