PardisSzah/FaithDPO-v1

Partial fine-tune vs allenai/Llama-3.1-Tulu-3-8B-SFT. Partial change: 84% of the weights changed (attention, mlp, norms), in 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 11/65, MLP 94/96, Attention 118/128, lm_head 0/1

1 derivatives checked

• mradermacher/FaithDPO-v1-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse