ZeroXClem/Stheno-Hercules-3.1-8B

Full fine-tune vs Locutusque/Hercules-6.1-Llama-3.1-8B. Fine-tune: 91% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 0/1, Norms 0/65, MLP 93/96, Attention 124/128, lm_head 1/1

3 derivatives checked

• bartowski/Stheno-Hercules-3.1-8B-GGUF: Same weights, converted
• mradermacher/Stheno-Hercules-3.1-8B-i1-GGUF: Quantization
• mradermacher/Stheno-Hercules-3.1-8B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse