TheBloke/13B-BlueMethod-GPTQ
Full fine-tune vs CalderaAI/13B-BlueMethod. Fine-tune: 91% of the weights changed, across 40 of 40 layers.
Tensors changed: MLP 120/240, Attention 160/360, lm_head 0/1, Embeddings 0/1, Norms 0/81
Full fine-tune vs CalderaAI/13B-BlueMethod. Fine-tune: 91% of the weights changed, across 40 of 40 layers.
Tensors changed: MLP 120/240, Attention 160/360, lm_head 0/1, Embeddings 0/1, Norms 0/81