Sakalti/SJT-2B-V1.1

Full fine-tune vs prithivMLmods/GWQ2b. Fine-tune: 100% of the weights changed, across 26 of 26 layers.

Tensors changed: Embeddings 1/1, Norms 105/105, MLP 78/78, Attention 104/104

2 derivatives checked

• mradermacher/SJT-2B-V1.1-GGUF: Same weights, converted
• mradermacher/SJT-2B-V1.1-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse