mindw96/Granite-3.1-8B-instruct-Korean

Full fine-tune vs ibm-granite/granite-3.1-8b-instruct. Fine-tune: 99% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 1/1, Norms 79/81, MLP 120/120, Attention 149/160

2 derivatives checked

• mradermacher/Granite-3.1-8B-instruct-Korean-GGUF: Same weights, converted
• mradermacher/Granite-3.1-8B-instruct-Korean-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse