mradermacher/saiga_gemma2_9b_abliterated_sft_m3_d9_abliterated-GGUF

Same weights, converted vs IlyaGusev/saiga_gemma2_9b_abliterated_sft_m3_d9_abliterated. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/169, MLP 0/126, Attention 0/168

On the Hub · Downloads on Model Pulse