Thao Nguyen

tmknguyen's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in matthewshu/medcasereasoning-cot-activations 3 days ago

Add screen1000 TVD raw activations

#11 opened 3 days ago by

tmknguyen

New activity in matthewshu/medcasereasoning-cot-activations about 1 month ago

Re-collect 2026 outline models' activations with corrected sentence splitter

#10 opened about 1 month ago by

tmknguyen

Add 2026 roster activations and re-extracted gradings

#9 opened about 1 month ago by

tmknguyen

add sae_topk_results for gemma-4-31b, qwen3.6-27b, glm-4.7-flash, ministral-3-14b-reasoning-2512

#3 opened 2 months ago by

tmknguyen

Add recommended_decoding activations for gemma-4-31b and glm-4.7-flash, and regrade both

1

#8 opened about 1 month ago by

tmknguyen

New activity in matthewshu/medcasereasoning-cot-activations about 2 months ago

Add recommended_decoding responses for gemma-4-31b, glm-4.7-flash, ministral-3-14b-reasoning-2512, qwen3.6-27b

#7 opened about 2 months ago by

tmknguyen

Add recommended_decoding data for gemma-4-31b, glm-4.7-flash, ministral-3-14b-reasoning-2512, qwen3.6-27b (JSON unicode escapes decoded to literal characters)

1

#6 opened about 2 months ago by

tmknguyen

Add recommended_decoding data for gemma-4-31b, glm-4.7-flash, ministral-3-14b-reasoning-2512, qwen3.6-27b (Ministral activations have <answer> emission sentences stripped)

1

#5 opened about 2 months ago by

tmknguyen

Add recommended_decoding data (responses/raw activations/SAEs) for gemma-4-31b, glm-4.7-flash, ministral-3-14b-reasoning-2512, qwen3.6-27b

1

#4 opened about 2 months ago by

tmknguyen

New activity in matthewshu/medcasereasoning-cot-activations 2 months ago

add gemma-4-31b, qwen3.6-27b, glm-4.7-flash, ministral-3-14b-reasoning-2512: responses, activations, saes

#2 opened 2 months ago by

tmknguyen