pszemraj/granite-4.0-h-7b-heretic

Partial fine-tune vs ibm-granite/granite-4.0-h-tiny. Partial change: 19% of the weights changed (attention, mlp, other), in 29 of 40 layers.

Tensors changed: Embeddings 0/1, other 50/372, Norms 0/117, MLP 20/80, Attention 1/16

2 derivatives checked

• mradermacher/granite-4.0-h-7b-heretic-i1-GGUF: Quantization
• mradermacher/granite-4.0-h-7b-heretic-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse