virtuous7373/Mira-M2

Full fine-tune vs Lambent/Mira-1.10-dpo-27B. Fine-tune: 99% of the weights changed, across 62 of 62 layers.

Tensors changed: Embeddings 1/4, Norms 141/484, MLP 197/294, Attention 271/464, other 1/1

3 derivatives checked

• mradermacher/Mira-M2-i1-GGUF: Quantization (fewer layers)
• mradermacher/Mira-M2-GGUF: Quantization (fewer layers)
• Lambent/Mira-v1.18-27B: LoRA-style

On the Hub · Downloads on Model Pulse