Sangwoo Park's picture

Sangwoo Park

Sangsang

·

swgger

AI & ML interests

I do LLM post-training research (KAIST AI)

Recent Activity

upvoted a paper 1 day ago

Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe

upvoted a paper 1 day ago

Memory Transfer Learning: How Memories are Transferred Across Domains in Coding Agents

updated a model 3 days ago

Sangsang/feedback_asymmetric_kl_fixed_ema_Qwen3-14B_bw0p5_fw0p5_ema0p999_ep30

View all activity

Organizations

None yet

Sangsang 's datasets 103

Sangsang/ThinkSafe-R1-Distill-8B-n5_filtered_Llama-3.3-70B-Instruct_eps10

Viewer • Updated Dec 24, 2025 • 154k • 4

Sangsang/ThinkSafe-8B-n5_filtered_Olmo-3.1-32B-Instruct_eps10

Viewer • Updated Dec 24, 2025 • 102k • 3

Sangsang/ThinkSafe-4B-n5_filtered_Olmo-3.1-32B-Instruct_eps10

Viewer • Updated Dec 24, 2025 • 97.5k • 3

Sangsang/ThinkSafe-R1-Distill-8B-n5_filtered_Olmo-3.1-32B-Instruct_eps10

Viewer • Updated Dec 24, 2025 • 77.1k • 3

Sangsang/ThinkSafe-0.6B-n5_filtered_Olmo-3.1-32B-Instruct_eps10

Viewer • Updated Dec 24, 2025 • 33k • 4

Sangsang/ThinkSafe-mixed-0.6B-8B

Viewer • Updated Dec 20, 2025 • 39.8k • 4

Sangsang/ThinkSafe-mixed-0.6B-4B

Viewer • Updated Dec 20, 2025 • 39.8k • 4

Sangsang/ThinkSafe-mixed-0.6B-1.7B

Viewer • Updated Dec 20, 2025 • 38.9k • 4

Sangsang/thinksafe_r1-distill_1.5B_force_2_harmful-refusal

Viewer • Updated Nov 26, 2025 • 34.9k • 4

Sangsang/thinksafe_r1-distill_1.5B_force_1_harmful-refusal

Viewer • Updated Nov 26, 2025 • 34.7k • 4

Sangsang/thinksafe_r1-distill_1.5B_force_2_harmful

Viewer • Updated Nov 26, 2025 • 38.6k • 4

Sangsang/thinksafe_r1-distill_1.5B_force_1_harmful

Viewer • Updated Nov 26, 2025 • 38.6k • 6

Sangsang/thinksafe_r1-distill_1.5B_forcing_3_debug

Viewer • Updated Nov 25, 2025 • 62 • 4

Sangsang/thinksafe_r1-distill_1.5B_forcing_2_debug

Viewer • Updated Nov 25, 2025 • 62 • 5

Sangsang/ThinkSafe-4B-n5-refusal_20000_5

Viewer • Updated Nov 22, 2025 • 235k • 5

Sangsang/ThinkSafe-0.6B-n5-refusal_20000_5

Viewer • Updated Nov 22, 2025 • 204k • 4

Sangsang/ThinkSafe-R1-Distill-7B-n5-refusal_20000_5

Viewer • Updated Nov 21, 2025 • 261k • 4

Sangsang/ThinkSafe-R1-Distill-1.5B-n5-refusal_20000_5

Viewer • Updated Nov 21, 2025 • 204k • 4

Sangsang/thinksafe-r1-distill-1.5B-n5_filtered_all_deepseekr-preview_5000_5

Viewer • Updated Nov 19, 2025 • 184k • 3

Sangsang/r1-8B_deepseekr-preview_5000_5

Viewer • Updated Nov 19, 2025 • 25k • 4

Sangsang/r1-7B_deepseekr-preview_5000_5

Viewer • Updated Nov 19, 2025 • 25k • 5

Sangsang/r1-1.5B_deepseekr-preview_5000_5

Viewer • Updated Nov 19, 2025 • 25k • 4

Sangsang/thinksafe-r1-distill-8B-n5-filtered-all

Viewer • Updated Nov 17, 2025 • 191k • 4

Sangsang/thinksafe-8B-n5-filtered-all

Viewer • Updated Nov 17, 2025 • 197k • 4

Sangsang/thinksafe-r1-distill-7B-n5_filtered_all

Viewer • Updated Nov 17, 2025 • 216k • 4

Sangsang/thinksafe-4B-n5-filtered-all

Viewer • Updated Nov 17, 2025 • 194k • 3

Sangsang/thinksafe-1.7B-n5_filtered_all

Viewer • Updated Nov 17, 2025 • 176k • 4

Sangsang/thinksafe-0.6B-n5_filtered_all

Viewer • Updated Nov 17, 2025 • 181k • 3

Sangsang/thinksafe-r1-distill-1.5B-n5_filtered_all

Viewer • Updated Nov 17, 2025 • 173k • 4

Sangsang/ThinkSafe-R1-Distill-1.5B-n5_math-n5-t3_8k

Viewer • Updated Nov 11, 2025 • 195k • 2