finetuned_codegen / tokenizer_config.json
anujsahani01's picture
Training in progress, epoch 0
9c83a36
raw
history blame
359 Bytes
{
"add_prefix_space": false,
"additional_special_tokens": [
"<|system|>",
"<|assistant|>",
"<|user|>",
"<|end|>",
"<|PAD|>"
],
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_max_length": 1400,
"tokenizer_class": "CodeGenTokenizer",
"unk_token": "<|endoftext|>"
}