crosloengual-bert / tokenizer_config.json
matejulcar's picture
added tokenizer_config, so the tokenizer is cased by default
67202cd
raw
history blame
46 Bytes
{
"do_lower_case": false,
"max_len": 512
}