A standard LLaMA-3.2 model trained with a pretrained Byte-Pair-Encoding (BPE) tokenizer. It provides a benchmark for evaluating the performance of more advanced tokenization approaches.
Downloads last month
2
Safetensors
Model size
1B params
Tensor type
F16
·
Collection including OpenFacedSandwich/STD-BPE-LLAMA3.2
