mradermacher/JAIS2-IT-0.3-GGUF
Same weights, converted vs yoriis/JAIS2-IT-0.3. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/130, MLP 0/128, Attention 0/256
Same weights, converted vs yoriis/JAIS2-IT-0.3. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/130, MLP 0/128, Attention 0/256