huihui-ai/Huihui-granite-4.0-micro-abliterated
LoRA-style vs ibm-granite/granite-4.0-micro. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of o_proj, output_linear changed, in 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/81, Attention 40/160, MLP 40/80
2 derivatives checked
- mradermacher/Huihui-granite-4.0-micro-abliterated-i1-GGUF: Quantization
- mradermacher/Huihui-granite-4.0-micro-abliterated-GGUF: Same weights, converted