sail/Sailor2-L-1B

Full fine-tune vs sail/Sailor2-1B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 336/336

3 derivatives checked

• mradermacher/Sailor2-L-1B-GGUF: Same weights, converted
• sail/Sailor2-L-1B-Chat: Full fine-tune
• sail/Sailor2-L-1B-SFT: Full fine-tune

On the Hub · Downloads on Model Pulse