classical-ja-tokenizer / verification.json
kuramot6f's picture
Initial release (history squashed)
5fe47a9
Raw History Blame Contribute Delete
1.84 kB
{
"passed": true,
"verified": "2026-10-20T10:05:10.712512+00:00",
"vocab_size": 32000,
"byte_pieces": 256,
"special_ids": {
"unk": 0,
"bos": 1,
"eos": 2,
"pad": 3
},
"by_source": {
"kanripo": {
"samples": 128,
"characters": 244506,
"tokens": 197796,
"unknown_tokens": 0,
"byte_tokens": 13491,
"roundtrip_matches": 128,
"characters_per_token": 1.236152399441849,
"byte_token_fraction": 0.0682066371412971
},
"minhon": {
"samples": 128,
"characters": 38835,
"tokens": 26888,
"unknown_tokens": 0,
"byte_tokens": 3017,
"roundtrip_matches": 128,
"characters_per_token": 1.4443246057720915,
"byte_token_fraction": 0.11220618863433501
},
"ndl": {
"samples": 128,
"characters": 213716,
"tokens": 156139,
"unknown_tokens": 0,
"byte_tokens": 677,
"roundtrip_matches": 128,
"characters_per_token": 1.368754763383908,
"byte_token_fraction": 0.004335880209300687
},
"wikisource": {
"samples": 128,
"characters": 219399,
"tokens": 155118,
"unknown_tokens": 0,
"byte_tokens": 3874,
"roundtrip_matches": 128,
"characters_per_token": 1.414400649827873,
"byte_token_fraction": 0.024974535514898335
},
"_probes": {
"samples": 9,
"characters": 102,
"tokens": 126,
"unknown_tokens": 0,
"byte_tokens": 57,
"roundtrip_matches": 9,
"characters_per_token": 0.8095238095238095,
"byte_token_fraction": 0.4523809523809524
}
},
"failures": [],
"native_u2581_roundtrip": false,
"required_wrapper": "scripts/sentencepiece_unigram.py:Tokenizer for literal U+2581",
"evaluation_limit": "Diagnostic training-distribution sample, not held-out evaluation"
}