mergekit-community/Roci-Maxx

Full fine-tune vs ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.1. Fine-tune: 97% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, Norms 25/65, MLP 93/96, Attention 124/128, lm_head 1/1

2 derivatives checked

• mradermacher/Roci-Maxx-i1-GGUF: Quantization
• mradermacher/Roci-Maxx-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse