Sakalti/Saba2-3B

Same weights, converted vs unsloth/Qwen2.5-3B-Instruct. Same weights stored in another dtype (BF16 → F16).

Tensors changed: Embeddings 0/1, Norms 0/73, MLP 0/108, Attention 0/252

1 derivatives checked

• mradermacher/Saba2-3B-GGUF: Same weights, converted

3 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse