google/t5gemma-b-b-ul2-it
Partial fine-tune vs google/t5gemma-b-b-ul2. Partial change: 67% of the weights changed (attention, embeddings, mlp, norms), in 12 of 12 layers.
Tensors changed: Embeddings 1/2, Attention 144/144, MLP 72/72, Norms 122/122
Partial fine-tune vs google/t5gemma-b-b-ul2. Partial change: 67% of the weights changed (attention, embeddings, mlp, norms), in 12 of 12 layers.
Tensors changed: Embeddings 1/2, Attention 144/144, MLP 72/72, Norms 122/122