Sangsang/ci-sft_Qwen3-4B
Text Generation • Updated
Sangwoo Park
Sangsang
AI & ML interests
I do LLM post-training research (KAIST AI)
Recent Activity
updated a model about 1 hour ago
Sangsang/ci-sft_Qwen3-4B published a model about 1 hour ago
Sangsang/ci-sft_Qwen3-4B updated a dataset about 3 hours ago
Sangsang/CI-Qwen3-32B-Instruct-Augmented-ResponsesOrganizations
None yet
Sangsang/ci-sft_Qwen3-4B-Instruct-2507_lr1e-6_ep30
Text Generation • Updated
Sangsang/qwen3_onpolicy
Text Generation • Updated • 30
Sangsang/grpo_Qwen3-1.7B_bs16_g16_mb128_lr1e-6_b1e-3_clip0p2_temp0p7_ep30
Text Generation • Updated • 11
Sangsang/grpo_Qwen3-4B_bs16_g16_mb128_lr1e-6_b1e-3_clip0p2_temp0p7_ep30
Text Generation • Updated • 15
Sangsang/feedback_asymmetric_fixed_ema_DeepSeek-R1-Distill-Qwen-7B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 14
Sangsang/feedback_asymmetric_fixed_ema_DeepSeek-R1-Distill-Llama-8B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 15
Sangsang/feedback_both_ema_Qwen3-4B_reverse_kl_ema0p999_ep30
Text Generation • Updated • 12
Sangsang/feedback_allowed_ema_Qwen3-4B_reverse_kl_ema0p999_ep30
Text Generation • Updated • 16
Sangsang/feedback_disallowed_ema_Qwen3-4B_reverse_kl_ema0p999_ep30
Text Generation • Updated • 12
Sangsang/feedback_asymmetric_fixed_ema_Qwen3-4B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 14
Sangsang/grpo_Qwen3-4B-Instruct-2507_bs16_g16_mb128_lr1e-6_b1e-3_clip0p2_temp0p7_ep30
Text Generation • Updated • 18
Sangsang/feedback_asymmetric_fixed_ema_Qwen3-4B-Instruct-2507_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 16
Sangsang/feedback_disallowed_ema_Qwen3-4B-Instruct-2507_reverse_kl_ema0p999_ep30
Text Generation • Updated • 23
Sangsang/feedback_both_ema_Qwen3-4B-Instruct-2507_reverse_kl_ema0p999_ep30
Text Generation • Updated • 13
Sangsang/feedback_allowed_ema_Qwen3-4B-Instruct-2507_reverse_kl_ema0p999_ep30
Text Generation • Updated • 13
Sangsang/feedback_asymmetric_fixed_ema_Qwen2.5-7B-Instruct_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 20
Sangsang/feedback_disallowed_ema_Qwen2.5-7B-Instruct_reverse_kl_ema0p999_ep30
Text Generation • Updated • 12
Sangsang/feedback_both_ema_Qwen2.5-7B-Instruct_reverse_kl_ema0p999_ep30
Text Generation • Updated • 12
Sangsang/feedback_allowed_ema_Qwen2.5-7B-Instruct_reverse_kl_ema0p999_ep30
Text Generation • Updated • 13
Sangsang/feedback_asymmetric_fixed_ema_Llama-3.1-8B-Instruct_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 27
Sangsang/feedback_both_ema_Llama-3.1-8B-Instruct_reverse_kl_ema0p999_ep30
Text Generation • Updated • 16
Sangsang/feedback_disallowed_ema_Llama-3.1-8B-Instruct_reverse_kl_ema0p999_ep30
Text Generation • Updated • 12
Sangsang/feedback_allowed_ema_Llama-3.1-8B-Instruct_reverse_kl_ema0p999_ep30
Text Generation • Updated • 18
Sangsang/Llama-3.1-8B-rkl_rkl-4130
Text Generation • Updated • 13
Sangsang/qwen3-8B-safepath-Qwen-8B-32-pm-3ep
Text Generation • Updated • 24
Sangsang/R1-8B-safepath-DeepSeek-8B-32-pm-3ep
Text Generation • Updated • 24
Sangsang/qwen3-8B-thinksafe-Qwen-8B-unfiltered-raw-32-pm-3ep
Text Generation • Updated • 25
Sangsang/Olmo-3-7B-Think_from_Olmo-3-32B-Think
Text Generation • Updated • 15
Sangsang/R1-8B-thinksafe-DeepSeek-8B-unfiltered-raw-32-pm-3ep
Text Generation • Updated • 22