Nexesenex/Llama_3.1_8b_Hermedive_R1_V1.01

Full fine-tune vs NousResearch/DeepHermes-3-Llama-3-8B-Preview. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

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

3 derivatives checked

• mradermacher/Llama_3.1_8b_Hermedive_R1_V1.01-i1-GGUF: Quantization
• mradermacher/Llama_3.1_8b_Hermedive_R1_V1.01-GGUF: Same weights, converted
• Nexesenex/Llama_3.1_8b_DoberWild_v2.02: Full fine-tune

On the Hub · Downloads on Model Pulse