LWT-2.5B-checkpoints / tokenizer_meta.json
fillay's picture
Upload 3 files
ba575cd verified
Raw History Blame Contribute Delete
297 Bytes
{
"vocab_size": 32768,
"requested_vocab_size": 32768,
"pad_token": "<pad>",
"eos_token": "<|endoftext|>",
"pad_id": 0,
"eos_id": 1,
"sample": "/home/gmachine/data/pretrain_packed/tokenizer_sample.jsonl",
"train_seconds": 154.5,
"backend": "huggingface-tokenizers-bytelevel-bpe"
}