Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2
Paper • 2512.22671 • Published • 1 •
1
Pere Martra
oopere
AI & ML interests
ML Research Engineer | LLM Optimization & Efficiency | Authoring “Rearchitecting LLMs” @Manning | 1.9K GitHub ⭐
Recent Activity
liked a model about 7 hours ago
oopere/Llama-FinSent-S liked a model about 20 hours ago
oopere/llama-3.2-3b-attn-drop-3 updated a collection about 20 hours ago
Interesting LLM PapersOrganizations
commented a
paper 3 months ago
Upload Martra_Figure_5_Prompt_Tuning.jpg
2
#5 opened about 2 years ago
by oopere