gpt-tamil / tokenizer_config.json
aaparajit02's picture
Training in progress, step 10
8053cc4
raw
history blame contribute delete
221 Bytes
{
"add_prefix_space": false,
"bos_token": "<|endoftext|>",
"eos_token": "<|endoftext|>",
"model_max_length": 1000000000000000019884624838656,
"tokenizer_class": "GPT2Tokenizer",
"unk_token": "<|endoftext|>"
}