general-preference/GPM-Gemma-2B
Partial fine-tune vs google/gemma-2b-it. Partial change: 79% of the weights changed (attention, mlp, norms), in 18 of 18 layers.
Tensors changed: Embeddings 0/1, Norms 35/37, MLP 54/54, Attention 72/72
Partial fine-tune vs google/gemma-2b-it. Partial change: 79% of the weights changed (attention, mlp, norms), in 18 of 18 layers.
Tensors changed: Embeddings 0/1, Norms 35/37, MLP 54/54, Attention 72/72