rinna/nekomata-7b-instruction

Full fine-tune vs rinna/nekomata-7b. Fine-tune: 92% of the weights changed, across 32 of 32 layers.

Tensors changed: Attention 96/96, Norms 62/65, MLP 96/96, Embeddings 0/1, lm_head 1/1

2 derivatives checked

• mradermacher/nekomata-7b-instruction-i1-GGUF: Quantization
• mradermacher/nekomata-7b-instruction-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse