mradermacher/gemma-3-4b-it-roleplay-tuned-v1-GGUF

Same weights, converted (fewer layers) vs Indexnusrefather/gemma-3-4b-it-roleplay-tuned-v1. mradermacher/gemma-3-4b-it-roleplay-tuned-v1-GGUF matches part of Indexnusrefather/gemma-3-4b-it-roleplay-tuned-v1 (444 of its 883 tensors; the rest of Indexnusrefather/gemma-3-4b-it-roleplay-tuned-v1 has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136

On the Hub · Downloads on Model Pulse