mergekit-community/Roci_Maxx_v2

Full fine-tune vs ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.1. Fine-tune: 97% of the weights changed, across 33 of 33 layers.

Tensors changed: Embeddings 1/1, Norms 26/67, MLP 96/99, Attention 128/132, lm_head 1/1

2 derivatives checked

• mradermacher/Roci_Maxx_v2-i1-GGUF: Quantization
• mradermacher/Roci_Maxx_v2-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse