Jacob Hoover Vigly

postylem's activity

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

New activity in vxef/spider2-snow-temperature-sweep 2 months ago

Add all-pairs execution eval (quoted arm) + EXPLAIN compile verdicts (both arms)

2

#3 opened 2 months ago by

postylem

New activity in samuki-hf/tool-use 2 months ago

Add duplicate-rollouts checker: 2 cells have duplicated (instance_id, sample) rows

3

#3 opened 2 months ago by

postylem

Add duplicate-rollouts checker: classifies doubled uploads vs sample-id collisions

#4 opened 2 months ago by

postylem

think-arm tool rollouts: the <think> block before each tool call is generated but not stored

1

#2 opened 2 months ago by

postylem

New activity in vxef/spider2-snow-temperature-sweep 3 months ago

Add first-pass execution eval (evaluations/)

#2 opened 3 months ago by

postylem

New activity in samuki-hf/thinking-rollouts 3 months ago

Add LCB got store: model outputs keyed by output_hash, 8KB-capped + hash-sorted for cheap lookups

#2 opened 3 months ago by

postylem

New activity in samuki-hf/temperature-sweep-data 3 months ago

Add LCB test cases (input/expected; 8KB-capped display cols + full cols) next to prompts

#3 opened 3 months ago by

postylem

Update README.md

#2 opened 3 months ago by

postylem