Ateron/Sketch-Cydonia-24B-V1.3

Partial fine-tune vs Ateron/Sketch-Cydonia-24B-V1.2. Partial change: 61% of the weights changed (attention, mlp, norms), in 26 of 40 layers.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 1/81, MLP 78/120, Attention 104/160

2 derivatives checked