automerger/Percival_01Yamshadow-7B
Full fine-tune vs AurelPx/Percival_01-7b-slerp. Fine-tune: 97% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 93/96, Attention 124/128, Norms 35/65, Embeddings 1/1, lm_head 1/1
1 derivatives checked
- MaziyarPanahi/Percival_01Yamshadow-7B-GGUF: Quantization