mradermacher/IBM_granite-4.1-3b_Abliterated-GGUF

Same weights, converted vs SicariusSicariiStuff/IBM_granite-4.1-3b_Abliterated. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/120, Attention 0/160

On the Hub · Downloads on Model Pulse