DFveloper's picture
download
raw
3.95 kB
{
"seq_len": 32768,
"long_seq_len": 65536,
"total_rows": 1568649,
"total_tokens": 1871250901,
"korean_rows": 1156544,
"long64k_rows": 4607,
"buckets": {
"korean": [
{
"pool": "merged",
"read": 3866990,
"kept": 1125921,
"role_prefix_stripped": 178865,
"tokens": 445000555,
"korean": 1125921,
"long64k": 21,
"stride": 1,
"dropped": {
"stub_answer": 30845,
"truncated": 1365,
"degenerate_repetition": 1155,
"contaminated": 1199,
"boilerplate_refusal": 1882,
"no_final_answer": 326,
"lang_tag_ko_no_hangul": 120,
"template_leak": 177,
"over_64k": 8,
"consecutive_assistant": 5
}
}
],
"core_en": [
{
"pool": "merged",
"read": 1337331,
"kept": 114399,
"role_prefix_stripped": 89418,
"tokens": 240002538,
"korean": 0,
"long64k": 0,
"stride": 1,
"dropped": {
"stub_answer": 1148,
"truncated": 451,
"degenerate_repetition": 279,
"contaminated": 235,
"boilerplate_refusal": 44,
"no_final_answer": 163,
"lang_tag_ko_no_hangul": 60,
"template_leak": 3
}
}
],
"tool": [
{
"pool": "merged",
"read": 2306564,
"kept": 262284,
"role_prefix_stripped": 4,
"tokens": 634327173,
"korean": 5107,
"long64k": 1676,
"stride": 1,
"dropped": {
"contaminated": 938,
"degenerate_repetition": 21,
"template_leak": 7,
"boilerplate_refusal": 751,
"row_cap": 616049,
"consecutive_assistant": 3323,
"no_final_answer": 1,
"truncated": 2,
"unbalanced_think": 382
}
}
],
"korean_cot": [
{
"pool": "ko_cot",
"read": 25515,
"kept": 25514,
"role_prefix_stripped": 0,
"tokens": 21907428,
"korean": 25514,
"long64k": 0,
"stride": 1,
"dropped": {
"degenerate_repetition": 1
}
}
],
"reasoning": [
{
"pool": "openthoughts3",
"read": 38535,
"kept": 13042,
"role_prefix_stripped": 0,
"tokens": 180005956,
"korean": 0,
"long64k": 0,
"stride": 12,
"dropped": {
"unbalanced_think": 25439,
"degenerate_repetition": 33,
"truncated": 21
}
}
],
"long_context": [
{
"pool": "longcite_full",
"read": 15034,
"kept": 6688,
"role_prefix_stripped": 0,
"tokens": 220003780,
"korean": 2,
"long64k": 2903,
"stride": 1,
"dropped": {
"over_64k": 1566,
"degenerate_repetition": 12
}
},
{
"pool": "longwriter",
"read": 1171,
"kept": 1170,
"role_prefix_stripped": 0,
"tokens": 10002342,
"korean": 0,
"long64k": 7,
"stride": 1,
"dropped": {
"stub_answer": 1
}
}
],
"code": [
{
"pool": "opencodereasoning",
"read": 9415,
"kept": 9414,
"role_prefix_stripped": 0,
"tokens": 80000195,
"korean": 0,
"long64k": 0,
"stride": 41,
"dropped": {
"unbalanced_think": 1
}
}
],
"reasoning_math": [
{
"pool": "openr1_math",
"read": 11738,
"kept": 10217,
"role_prefix_stripped": 0,
"tokens": 40000934,
"korean": 0,
"long64k": 0,
"stride": 5,
"dropped": {
"math_answer_wrong": 1498,
"unbalanced_think": 11,
"degenerate_repetition": 12
}
}
]
}
}

Xet Storage Details

Size:
3.95 kB
·
Xet hash:
1d733d0fb9a2191078bc5b35ed385db608da2c7b98581d04b28efe840c2eb65e

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.