Nexesenex/Llama_3.2_3b_Shark_0.0

Full fine-tune vs SaisExperiments/Evil-Alpaca-3B-L3.2. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Norms 57/57, MLP 84/84, Attention 112/112

2 derivatives checked

• mradermacher/Llama_3.2_3b_Shark_0.0-i1-GGUF: Quantization
• mradermacher/Llama_3.2_3b_Shark_0.0-GGUF: LoRA-style

On the Hub · Downloads on Model Pulse