pszemraj/rnj-1.5-instruct
Full fine-tune vs EssentialAI/rnj-1. Fine-tune: 98% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 193/193, MLP 94/96, Attention 128/128
1 derivatives checked
- pszemraj/rnj-1.5-instruct-GGUF: Quantization