Hastagaras/L3.2-JametMini-3B-MK.III
Full fine-tune vs meta-llama/Llama-3.2-3B-Instruct. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 15/57, MLP 84/84, Attention 112/112
5 derivatives checked
- mradermacher/L3.2-JametMini-3B-MK.III-GGUF: Same weights, converted
- mradermacher/L3.2-JametMini-3B-MK.III-i1-GGUF: Quantization
- Xiaojian9992024/Reflection-L3.2-JametMiniMix-3B: Full fine-tune
- mergekit-community/JAJUKA-WEWILLNEVERFORGETYOU-3B: Full fine-tune
- bunnycore/Llama-3.2-3B-Stock: Full fine-tune