Lambent/Zora-9B-v2

Full fine-tune vs Lambent/Zora-9B-v1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: lm_head 1/1, Embeddings 5/5, Norms 216/221, Attention 326/336, MLP 211/211, other 1/1

2 derivatives checked

• mradermacher/Zora-9B-v2-i1-GGUF: Quantization
• mradermacher/Zora-9B-v2-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse