mradermacher/Olmo-3-7B-Instruct-OPSA-GGUF
Same weights, converted vs Tuwhy/Olmo-3-7B-Instruct-OPSA. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, MLP 0/96, Attention 0/128, lm_head 0/1, Norms 0/129
Same weights, converted vs Tuwhy/Olmo-3-7B-Instruct-OPSA. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, MLP 0/96, Attention 0/128, lm_head 0/1, Norms 0/129