Khetterman/Llama-3.2-Kapusta-JapanChibi-3B-v1
Full fine-tune vs AELLM/Llama-3.2-Chibi-3B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112
2 derivatives checked
- mradermacher/Llama-3.2-Kapusta-JapanChibi-3B-v1-i1-GGUF: Quantization
- mradermacher/Llama-3.2-Kapusta-JapanChibi-3B-v1-GGUF: Same weights, converted