mlabonne/Hermes-3-Llama-3.1-70B-lorablated

Partial fine-tune vs NousResearch/Hermes-3-Llama-3.1-70B. Partial change: 45% of the weights changed (attention, mlp, norms), in 80 of 80 layers.

Tensors changed: Embeddings 0/1, Norms 10/161, MLP 108/240, Attention 116/320, lm_head 0/1

4 derivatives checked