LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B
Paper
•
2310.20624
•
Published
•
13
Interesting ones
-
-
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
Paper • 2310.20587 • Published • 18 -
BadLlama: cheaply removing safety fine-tuning from Llama 2-Chat 13B
Paper • 2311.00117 • Published -
VideoFusion: Decomposed Diffusion Models for High-Quality Video Generation
Paper • 2303.08320 • Published • 3 -
Vikhrmodels/Vikhr-7B-instruct_0.4
Text Generation • Updated • 5.84k • 31 -
IlyaGusev/saiga_llama3_8b
Text Generation • Updated • 9.75k • 119 -
cognitivecomputations/wizard_vicuna_70k_unfiltered
Viewer • Updated • 34.6k • 137 • 161 -
failspy/llama-3-70B-Instruct-abliterated
Text Generation • Updated • 5.52k • 102 -
Zoyd/Sao10K_L3-8B-Stheno-v3.1-8_0bpw_exl2
Text Generation • Updated • 8 • 3 -
Zoyd/Sao10K_L3-8B-Stheno-v3.1-6_5bpw_exl2
Text Generation • Updated • 10 • 1 -
sophosympatheia/Aurora-Nights-70B-v1.0
Text Generation • Updated • 1.4k • 22 -
PygmalionAI/mythalion-13b
Text Generation • Updated • 2.67k • 158 -
Nitral-AI/Poppy_Porpoise-1.0-L3-8B
Text Generation • Updated • 27 • 24 -
NeverSleep/Noromaid-v0.4-Mixtral-Instruct-8x7b-Zloss
Text Generation • Updated • 82 • 36 -
microsoft/Phi-3-medium-128k-instruct
Text Generation • Updated • 33.2k • 378 -
Azazelle/L3-RP_io
Text Generation • Updated • 165 • 3 -
Lewdiculous/Poppy_Porpoise-1.0-L3-8B-GGUF-IQ-Imatrix
Updated • 182 • 15 -
ACECODER: Acing Coder RL via Automated Test-Case Synthesis
Paper • 2502.01718 • Published • 22