Sakalti/SJT-2.1B

Same weights, repacked vs Sakalti/SJT-1.5B-Alpha. Same weights in different files (re-saved, re-sharded or new metadata), plus 84 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 0/196

2 derivatives checked

• mradermacher/SJT-2.1B-i1-GGUF: Quantization
• mradermacher/SJT-2.1B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse