Mungert/granite-4.0-h-350m-GGUF
Full fine-tune vs ibm-granite/granite-4.0-h-350m-base. Mungert/granite-4.0-h-350m-GGUF is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 61/65, Attention 16/16