nlpguy/Mistral-NeMo-Minitron-Upscale-v2
Partial fine-tune (fewer layers) vs nvidia/Mistral-NeMo-Minitron-8B-Base. nvidia/Mistral-NeMo-Minitron-8B-Base matches part of nlpguy/Mistral-NeMo-Minitron-Upscale-v2 (363 of its 561 tensors; the rest of nlpguy/Mistral-NeMo-Minitron-Upscale-v2 has no counterpart). On the shared part: Partial change: 60% of the weights changed (attention, mlp, norms), in 32 of 40 layers.
Tensors changed: Embeddings 0/1, Norms 48/81, MLP 80/120, Attention 120/160, lm_head 0/1
2 derivatives checked
- mradermacher/Mistral-NeMo-Minitron-Upscale-v2-i1-GGUF: Quantization
- mradermacher/Mistral-NeMo-Minitron-Upscale-v2-GGUF: Quantization