sail/Sailor-4B

Full fine-tune vs Qwen/Qwen1.5-4B. Fine-tune: 100% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 1/1, Norms 81/81, MLP 120/120, Attention 280/280, lm_head 1/1

4 derivatives checked

• mradermacher/Sailor-4B-GGUF: Same weights, converted
• sail/Sailor-4B-Chat-gguf: Full fine-tune
• sail/Sailor-4B-Chat: Full fine-tune
• Sakalti/anchobi-4b: Full fine-tune

On the Hub · Downloads on Model Pulse