shisa-ai/shisa-v2-mistral-nemo-12b
Full fine-tune vs mistralai/Mistral-Nemo-Instruct-2407. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 1/1, Norms 36/81, MLP 120/120, Attention 160/160, lm_head 1/1
6 derivatives checked
- mradermacher/shisa-v2-mistral-nemo-12b-i1-GGUF: Quantization
- mradermacher/shisa-v2-mistral-nemo-12b-GGUF: Quantization
- DevQuasar/shisa-ai.shisa-v2-mistral-nemo-12b-GGUF: Same weights, converted
- yamatazen/HMS-Fusion-12B: Partial fine-tune
- mpasila/shisa-v2-JP-EN-Translator-v0.1-12B: LoRA-style
- yamatazen/HMS-Slerp-12B: Full fine-tune
1 other repos hold byte-for-byte identical weight files.