ibm-research/merlinite-7b
Full fine-tune vs mistralai/Mistral-7B-v0.1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 18/65, MLP 96/96, Attention 128/128
5 derivatives checked
- mradermacher/merlinite-7b-i1-GGUF: Quantization
- MaziyarPanahi/merlinite-7b-GGUF: Quantization
- mradermacher/merlinite-7b-GGUF: Same weights, converted
- tensorblock/merlinite-7b-GGUF: Quantization
- StanFurrer/merlinite-7B-slerp: Full fine-tune
1 other repos hold byte-for-byte identical weight files.