PyLoomer_CodeParrot / tokenizer_config.json
anujsahani01's picture
Training in progress, step 7500
2064e14
raw
history blame
385 Bytes
{
"add_prefix_space": false,
"additional_special_tokens": [
"<SYSTEM_TASK:>",
"<USER_TASK:>",
"<ASSISTANT_TASK:>",
"<END_TASK>"
],
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_max_length": 2048,
"pad_token": "<|PAD|>",
"tokenizer_class": "CodeGenTokenizer",
"unk_token": "<|endoftext|>"
}