sail/Sailor2-1B-SFT
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 33/97, MLP 144/144, Attention 333/336
3 derivatives checked
- mradermacher/Sailor2-1B-SFT-i1-GGUF: Quantization
- mradermacher/Sailor2-1B-SFT-GGUF: Same weights, converted
- Sakalti/Saka-1B: Same weights, converted