Separate the Wheat from the Chaff: Model Deficiency Unlearning via Parameter-Efficient Module Operation
Paper • 2308.08090 • Published
Ext-Sub
arxiv.org/abs/2308.08090
-
-
Ext-Sub/llama-13b_alpaca-gpt4_lora
Updated -
Ext-Sub/llama-7b_wizardlm_untruthful_lora
Updated -
Ext-Sub/llama-13b_toxic_lora
Updated -
Ext-Sub/llama-7b_wizardlm_lora
Updated -
Ext-Sub/llama-7b_toxic_lora
Updated -
Ext-Sub/llama-13b_alpaca-gpt4-untruthful_lora
Updated -
Ext-Sub/llama-7b_alpaca-gpt4_lora
Updated -
Ext-Sub/llama-7b_alpaca-gpt4-untruthful_lora
Updated -
Ext-Sub/toxic_instruct
Preview • Updated • 10 -
Ext-Sub/wizardlm-70k-untruthful
Viewer • Updated • 55k • 593 -
Ext-Sub/alpaca-gpt4
Viewer • Updated • 52k • 14 -
Ext-Sub/alpaca-gpt4-untruthful
Viewer • Updated • 52k • 14 -
Ext-Sub/wizardlm-70k
Viewer • Updated • 55k • 506