rasyosef/Llama-3.2-1B-Amharic

Full fine-tune vs meta-llama/Llama-3.2-1B-Instruct. Fine-tune: 100% of the weights changed, across 16 of 16 layers.

Tensors changed: Norms 9/33, MLP 48/48, Attention 64/64

2 derivatives checked

• rasyosef/Llama-3.2-1B-Amharic-Instruct: LoRA-style
• mradermacher/Llama-3.2-1B-Amharic-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse