CaioXapelaum/Hermes-2-Pro-Llama-3.1-70B

Full fine-tune vs NousResearch/Hermes-2-Pro-Llama-3-70B. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: Norms 161/161, MLP 240/240, Attention 320/320

2 derivatives checked

• mradermacher/Hermes-2-Pro-Llama-3.1-70B-i1-GGUF: Quantization
• mradermacher/Hermes-2-Pro-Llama-3.1-70B-GGUF: Quantization

On the Hub · Downloads on Model Pulse