JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_best_at_epoch14 Updated Oct 31, 2025 • 1
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch17 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch16 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch15 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch14 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch13 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch12 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch11 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch10 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch9 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch8 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch7 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch6 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch5 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch4 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch3 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch2 Updated Oct 31, 2025
JingweiNi/uhead_claim_Qwen3-8B_fixed_prm_layer1_dim512_head16_e20_lr5e-4_pos3_epoch1 Updated Oct 31, 2025