20NG_raw_ELECTRA_5E / tokenizer_config.json
pig4431's picture
Training in progress, epoch 1
1f80928
raw
history blame contribute delete
410 Bytes
{
"cls_token": "[CLS]",
"do_lower_case": true,
"mask_token": "[MASK]",
"model_max_length": 512,
"name_or_path": "google/electra-small-discriminator",
"num_labels": 20,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"special_tokens_map_file": null,
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "ElectraTokenizer",
"truncation": true,
"unk_token": "[UNK]"
}