RedHatAI/granite-3.2-2b-instruct

Full fine-tune vs ibm-granite/granite-3.1-2b-instruct. Fine-tune: 100% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 1/1, Norms 59/81, MLP 120/120, Attention 160/160

2 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse