mradermacher/SJT-900M-GGUF

Same weights, converted vs Sakalti/SJT-900M. Same weights stored in another dtype (F16 → F16/F32).

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/85, MLP 0/126, Attention 0/294

On the Hub · Downloads on Model Pulse