rasyosef/Phi-1_5-Instruct-v0.1
Full fine-tune vs microsoft/phi-1_5. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: Norms 27/50, MLP 96/96, Attention 192/192
1 derivatives checked
- Felladrin/gguf-Q5_K_M-Phi-1_5-Instruct-v0.1: Quantization