sbintuitions/sarashina2.2-tts
Full fine-tune vs sbintuitions/sarashina2.2-0.5b-instruct-v0.1. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Norms 49/49, MLP 72/72, Attention 96/96
1 derivatives checked
- orsele/sarashina2.2-tts-gguf: Same weights, converted