automerger/Strangemerges_32Shadow-7B

Full fine-tune vs CorticalStack/shadow-clown-7B-slerp. Fine-tune: 94% of the weights changed, across 31 of 32 layers.

Tensors changed: MLP 90/96, Attention 120/128, Norms 12/65, Embeddings 1/1, lm_head 1/1

2 derivatives checked

• MaziyarPanahi/Strangemerges_32Shadow-7B-GGUF: Quantization
• MaziyarPanahi/YamshadowInex12_Strangemerges_32Shadow: Full fine-tune

On the Hub · Downloads on Model Pulse