tbilisi-ai-lab/kona2-12B-Instruct

Full fine-tune vs tbilisi-ai-lab/kona2-12B-Base. Fine-tune: 100% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 1/1, Norms 0/81, MLP 120/120, Attention 160/160, lm_head 1/1

3 derivatives checked

• mradermacher/kona2-12B-Instruct-i1-GGUF: Quantization
• tbilisi-ai-lab/kona2-12B: Full fine-tune
• mradermacher/kona2-12B-Instruct-GGUF: Quantization

On the Hub · Downloads on Model Pulse