nicholasKluge/Aira-2-124M-DPO

Full fine-tune vs nicholasKluge/Aira-2-124M. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: Attention 48/48, Norms 50/50, MLP 48/48, Embeddings 2/2

3 derivatives checked

• mradermacher/Aira-2-124M-DPO-i1-GGUF: Quantization
• mradermacher/Aira-2-124M-DPO-GGUF: Same weights, converted
• Felladrin/gguf-sharded-Aira-2-124M-DPO: Quantization

On the Hub · Downloads on Model Pulse