Nexesenex/Llama_3.2_3b_Kostume_v1

Full fine-tune vs Nexesenex/Llama_3.2_3b_SmartiCoatz_0.1b. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 17/57, MLP 84/84, Attention 112/112

1 derivatives checked

• mradermacher/Llama_3.2_3b_Costume_RC2_bf16-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse