mayflowergmbh/occiglot-7b-de-es-fr-it-en-instruct-dareties
Full fine-tune vs mistralai/Mistral-7B-v0.1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/occiglot-7b-de-es-fr-it-en-instruct-dareties-i1-GGUF: Quantization
- mradermacher/occiglot-7b-de-es-fr-it-en-instruct-dareties-GGUF: Same weights, converted