falcon-rewrite-581-smallset / tokenizer_config.json
sherryzha's picture
Training in progress, step 10
8e83abf
raw
history blame
248 Bytes
{
"add_prefix_space": false,
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_input_names": [
"input_ids",
"attention_mask"
],
"model_max_length": 2048,
"tokenizer_class": "PreTrainedTokenizerFast"
}