mradermacher/kanana-1.5-2.1b-base-GGUF
Same weights, converted vs OpenLLM-Korea/kanana-1.5-2.1b-base. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 0/128
Same weights, converted vs OpenLLM-Korea/kanana-1.5-2.1b-base. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 0/128