test_pretrain_gpt_pipeline / tokenizer_config.json
vietnhatthai's picture
End of training
77ef444
raw
history blame contribute delete
137 Bytes
{
"clean_up_tokenization_spaces": true,
"model_max_length": 512,
"tokenizer_class": "OpenAIGPTTokenizer",
"unk_token": "<unk>"
}