Ateron/Sketch-Cydonia-24B-V1.2
Full fine-tune vs Ateron/Sketch-Cydonia_V1.1. Fine-tune: 92% of the weights changed, across 39 of 40 layers.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 3/81, MLP 117/120, Attention 156/160
6 derivatives checked
- mradermacher/Sketch-Cydonia-24B-V1.2-i1-GGUF: Quantization
- mradermacher/Sketch-Cydonia-24B-V1.2-GGUF: Quantization
- McG-221/Sketch-Cydonia-24B-V1.2-mlx-8Bit: Quantization
- Ateron/Sketch-Cydonia-24B-V1.3: Partial fine-tune
- Sorihon/Sketch-Bereaved-Harbinger-24B: Full fine-tune
- NeuR0mancR/Pale-Sonneteer: Full fine-tune