bart-base-re-attention-seq-512 / tokenizer_config.json
GuysTrans's picture
Training in progress, step 500
784c5be
raw
history blame contribute delete
457 Bytes
{
"add_prefix_space": false,
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"cls_token": "<s>",
"eos_token": "</s>",
"errors": "replace",
"mask_token": "<mask>",
"max_length": 512,
"model_max_length": 1024,
"pad_token": "<pad>",
"sep_token": "</s>",
"stride": 0,
"tokenizer_class": "BartTokenizer",
"trim_offsets": true,
"truncation_side": "right",
"truncation_strategy": "longest_first",
"unk_token": "<unk>"
}