P0u4a/t0-mt-3b-base-philosophy-sdf
Full fine-tune vs dlab-spp/t0-mt-3b-base. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112
Full fine-tune vs dlab-spp/t0-mt-3b-base. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112