JoaoBoer/tofu_Llama-3.1-8B-Instruct_forget01_SatImp

Full fine-tune vs open-unlearning/tofu_Llama-3.1-8B-Instruct_full. Fine-tune: 93% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 11/65, MLP 96/96, Attention 128/128, lm_head 1/1

On the Hub · Downloads on Model Pulse