DyT-NoNorm-LLMs-REWILD
PublicReplacing LayerNorm with Dynamic Tanh (DyT) in DistilGPT2 + LoRA, evaluated on RE-WILD, Alpaca, and ShareGPT.
Creat:2025-05-10T13:48:32
Update:2025-06-17T12:15:55
0
Stars
0
Stars Increase
Replacing LayerNorm with Dynamic Tanh (DyT) in DistilGPT2 + LoRA, evaluated on RE-WILD, Alpaca, and ShareGPT.