Lexsi/llama32-3b-dolly-sft-drift
Text Generation • 3B • Updated • 2
AI & ML interests
Frontier research around Safe and aligned intelligence
Recent Activity
View all activity
Papers
Forgetting That Sticks: Quantization-Permanent Unlearning via Circuit Attribution
$C$-$ΔΘ$: Circuit-Restricted Weight Arithmetic for Selective Refusal
Lexsi/llama32-3b-code-sft-drift
Text Generation • 3B • Updated • 1
Lexsi/llama31-8b-medical-sft-drift
Text Generation • 8B • Updated • 4
Lexsi/llama31-8b-legal-sft-drift
Text Generation • 8B • Updated • 2
Lexsi/llama31-8b-hh-rlhf-aligned
Text Generation • 8B • Updated • 1
Lexsi/llama31-8b-gsm8k-sft-drift
Text Generation • 8B • Updated • 4
Lexsi/llama31-8b-dolly-sft-drift
Text Generation • 8B • Updated • 2
Lexsi/llama31-8b-code-sft-drift
Text Generation • 8B • Updated • 2
Lexsi/gemma3-4b-hh-rlhf-aligned
Image-Text-to-Text • 4B • Updated • 2
Lexsi/gemma3-4b-gsm8k-sft-drift
Image-Text-to-Text • 4B • Updated • 1
Lexsi/gemma3-4b-dolly-sft-drift
Image-Text-to-Text • 4B • Updated • 2
Lexsi/gemma3-4b-code-sft-drift
Image-Text-to-Text • 4B • Updated • 1
Lexsi/llama32-3b-hh-rlhf-aligned
Text Generation • 3B • Updated • 9
Lexsi/obl-ex-4b-curated
Text Generation • Updated • 9
Lexsi/obl-ex-4b-raw
Text Generation • Updated • 8
Lexsi/Orion-MSP
Updated • 77 • 10
Lexsi/Orion-BiX
Updated • 20 • 2