NousResearch/DeepHermes-3-Mistral-24B-Preview
Full fine-tune vs mistralai/Mistral-Small-24B-Base-2501. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 6/81, MLP 120/120, Attention 160/160
5 derivatives checked
- bartowski/NousResearch_DeepHermes-3-Mistral-24B-Preview-GGUF: Same weights, converted
- mradermacher/DeepHermes-3-Mistral-24B-Preview-i1-GGUF: Quantization
- mradermacher/DeepHermes-3-Mistral-24B-Preview-GGUF: Quantization
- DevQuasar-10/NousResearch.DeepHermes-3-Mistral-24B-Preview-GGUF: Same weights, converted
- AlSamCur123/DeepHermes-3-Mistral-24BContinuedFine: LoRA-style
3 other repos hold byte-for-byte identical weight files.