double7/Llama-2-7b-EAX

Full fine-tune vs meta-llama/Llama-2-7b-hf. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 58/65, MLP 96/96, Attention 128/128

2 derivatives checked

• mradermacher/Llama-2-7b-EAX-i1-GGUF: Quantization
• mradermacher/Llama-2-7b-EAX-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse