cicero / tokenizer_config.json
gmmeyer's picture
Cicero LLM: curriculum-tuned 111M Latin model (ONNX + PyTorch + tokenizer)
94d35d6 verified
raw
history blame contribute delete
168 Bytes
{
"tokenizer_class": "PreTrainedTokenizerFast",
"bos_token": "<s>",
"eos_token": "</s>",
"unk_token": "<unk>",
"pad_token": null,
"model_max_length": 2048
}