File size: 263 Bytes
ca9b276 | 1 2 3 4 5 6 7 8 9 10 11 | {
"tokenizer": "atom-fractus-v1",
"vocab_size": 266,
"max_span_bytes": 32,
"pack_mode": "linguistic",
"docs": 2,
"tokens": 1059743,
"bin": "atom_corpus.i16",
"dtype": "int16",
"note": "fresh Atom id stream. Do not mix with GPT-2 50257 shards."
} |