Built on the LLaMA-3.2 architecture, this model uses a standard Byte-Pair-Encoding (BPE) tokenizer optimized for Danish. It serves as a baseline for evaluating the performance of subword tokenization.
Downloads last month
4
Safetensors
Model size
1B params
Tensor type
F16
·
Collection including OpenFacedSandwich/DA-BPE-LLAMA3.2
