Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
Job has been terminated due to a temporary spike in resource usage and may be restarted later.
Error code:   JobManagerCrashedError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

assistant_tokens
int64
code_language
string
difficulty
float64
difficulty_basis
string
domain
string
final_answer
string
has_reference
bool
has_tests
bool
messages
list
problem_cluster_id
string
problem_id
string
problem_token_count
int64
provenance
dict
record_id
string
split
string
tests
unknown
token_count
int64
validation_evidence
list
validation_level
string
2,430
null
0.66
source_metadata
math
D
true
false
[ { "content": "If $\\lg 2$, $\\lg (2^{x}-1)$, $\\lg (2^{x}+3)$ form an arithmetic sequence, then the value of $x$ is ( ).\n\nA: $1$\n\nB: $0$ or $32$\n\nC: $32$\n\nD: $\\log_{2}5$", "role": "user" }, { "content": "<think>\nOkay, let's see. The problem says that lg 2, lg(2^x -1), lg(2^x +3) form an ar...
75962d4cbbde51fbe298d47aaf31946a860acf8ef1d12e78d9444ee387801b8e
75962d4cbbde51fbe298d47aaf31946a860acf8ef1d12e78d9444ee387801b8e
75
{ "config": "all", "dataset": "open-r1/OpenR1-Math-220k", "license": "unknown", "processing_version": "unify-v2-sha256-provenance", "row_index": 55628, "source_complete": true, "split": "train", "synthetic": false }
00001328aa871c381157f3513b6dbb4ecacacdf93d7d9f9503e4037dfac4ba93
train
null
2,514
[ { "deterministic": true, "evidence_digest": "8eaf21ded90a6ca1c7e4633e45b28f8eb8d47c21473ce4285986d32731f63dfd", "input_digest": "a5b78d6efb69b470d0123926b088d0e8f2cf01315057dc04d0b601a3dc90d1a2", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
answer_verified
3,187
null
0.5
source_metadata
instruction
null
false
false
[ { "content": "Please focus on the functionality of this problem and provide code in Objective-C:\nGiven a string `s` and a character `letter`, return _the **percentage** of characters in_ `s` _that equal_ `letter` _**rounded down** to the nearest whole percent._\n\n**Example 1:**\n\n**Input:** s = \"foobar \",...
b3ffa0fdb218dab9fa8ddbfe52f1d6219a16b358ae4023aee9b8df91df164479
b3ffa0fdb218dab9fa8ddbfe52f1d6219a16b358ae4023aee9b8df91df164479
234
{ "config": "reasoning-deepseek", "dataset": "QuixiAI/dolphin-r1", "license": "unknown", "processing_version": "unify-v2-sha256-provenance", "row_index": 158983, "source_complete": true, "split": "train", "synthetic": false }
000013e6325f16e08ab7f32cfe8b1dfddc5bcc515d978fc556ef9eae791d3e76
train
null
3,430
[ { "deterministic": true, "evidence_digest": "fe438c87dca7158cba855177c96c6ac634c045b3efebb245f996fe38e9eb42e6", "input_digest": "6b105e29076c77a4078969c245fa900c4dbcd1666191bfe94e417e051928e2bd", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
source_supervised
5,892
null
0.66
source_metadata
math
532
true
false
[ { "content": "Two three-letter strings, $aaa$ and $bbb$, are transmitted electronically. Each string is sent letter by letter. Due to faulty equipment, each of the six letters has a 1/3 chance of being received incorrectly, as an $a$ when it should have been a $b$, or as a $b$ when it should be an $a$. However,...
cbcf8646d4c8b7cbc1f3a76c19dbba0e536a61b2491fd6db666a9c5eb748d856
cbcf8646d4c8b7cbc1f3a76c19dbba0e536a61b2491fd6db666a9c5eb748d856
171
{ "config": "all", "dataset": "open-r1/OpenR1-Math-220k", "license": "unknown", "processing_version": "unify-v2-sha256-provenance", "row_index": 64620, "source_complete": true, "split": "train", "synthetic": false }
000014090987280c02f03b46ae79d77326c84f8bab1d64e6df8b974e801cf074
train
null
6,072
[ { "deterministic": true, "evidence_digest": "5d9945ba58f13a4bac47d8e34e28bb5386eeda0dcebf05bf16f6bc48d74fbacb", "input_digest": "0b892898580402846abcf648a8122775bd3a86d839847861b958e1d8e5f1e0aa", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
answer_verified
4,867
null
0.66
source_metadata
math
\sqrt{3}-1
true
false
[ { "content": "9. Given $\\cos (\\alpha+\\beta)=\\cos \\alpha+\\cos \\beta$, find the maximum value of $\\cos \\alpha$.\n\nTranslate the above text into English, please keep the original text's line breaks and format, and output the translation result directly.", "role": "user" }, { "content": "<thin...
e20b3f0a06d3bfe0c565343b87d12fc3f6c500359180cfcb94c0e7300ffe1883
e20b3f0a06d3bfe0c565343b87d12fc3f6c500359180cfcb94c0e7300ffe1883
55
{ "config": "all", "dataset": "open-r1/OpenR1-Math-220k", "license": "unknown", "processing_version": "unify-v2-sha256-provenance", "row_index": 92854, "source_complete": true, "split": "train", "synthetic": false }
0000148c5f368ae1e9bce2e6739a03bc5e3ad6c47800ab2db453319fb3dc95c1
train
null
4,931
[ { "deterministic": true, "evidence_digest": "5f3882bac5aced2ea5ac57b8f2c7004cdd37a71dea202245b6adcb64664e799c", "input_digest": "b682826721b42c75efe6b25069fd129b9f8b2b65aa19e64118528d4e925d6a2e", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
answer_verified
13,917
null
0.590605
problem_length_proxy
math
null
false
false
[ { "content": "Let $y=f(x)$ be a function defined on an interval $(-\\infty; a)\\cup (a;+\\infty)$. Consider a function defined as $g(x)=f(x-a)$. It is given that for every $x$, $g(x)$ is identical to its inverse, $g^{-1}(x)$. The sum of the limits of function $f$ when $x$ tends to $a$ from the left and from the...
690e3e250c9f45cb4b10aabf3b48cfee7ac945a36dd94c655fcd317e638fc11d
690e3e250c9f45cb4b10aabf3b48cfee7ac945a36dd94c655fcd317e638fc11d
135
{ "config": "default", "dataset": "open-thoughts/OpenThoughts3-1.2M", "license": "unknown", "processing_version": "unify-v2-sha256-provenance", "row_index": 691711, "source_complete": true, "split": "train", "synthetic": false }
00001babd1096a59ff884131c173c6db1fe544620b59b372d18ce5a9e393bfab
train
null
14,061
[ { "deterministic": true, "evidence_digest": "d67e6860a130ca2902e0eb6005af21576cb46cce6c4f205e24045eddab5e8cff", "input_digest": "14885cb89da7a6c750a7583e1152a564cddbba20e7f43fcc867e3f1ab5074ccf", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
source_supervised
4,104
null
0.7
source_metadata
math
null
false
false
[ { "content": "Let \\( x \\) be the real positive number that satisfies the equation \\( x^2 - 17x - 16\\sqrt{x} = 0 \\). Find the integer number \\( n \\) such that \\( x(n - x) \\) is an integer.", "role": "user" }, { "content": "<think>Okay, so I need to find a positive real number x that satisfie...
894dbad115c19e6199e04d390b4de976a943bdfe147b15539d993ad261725fe0
894dbad115c19e6199e04d390b4de976a943bdfe147b15539d993ad261725fe0
58
{ "config": "default", "dataset": "nvidia/OpenMathReasoning", "license": "unknown", "processing_version": "unify-v2-sha256-provenance", "row_index": 665523, "source_complete": true, "split": "tir", "synthetic": false }
00001feec10723d181a23c7c03a9b77e00424bc5bfbc55dfa91bd7e445ef5d01
train
null
4,171
[ { "deterministic": true, "evidence_digest": "fe9bef191342c8c0f715c8b6a54f503d21d0dec8bf1149ead040ceef5134a018", "input_digest": "e74cabd786f39b75da9f44b046febc624ed216e27b83684d97daf6e08b5ef29d", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
source_supervised
4,343
cpp
0.71
source_metadata
code
null
false
true
[ { "content": "Recently, Chef got obsessed with piano. He is a just a rookie in this stuff and can not move his fingers from one key to other fast enough. He discovered that the best way to train finger speed is to play scales.\nThere are different kinds of scales which are divided on the basis of their interval...
02ed5f1879b81f981c789eaa8555b88a67adfd082378b99ce726e94be0101007
02ed5f1879b81f981c789eaa8555b88a67adfd082378b99ce726e94be0101007
780
{ "config": "train", "dataset": "nvidia/OpenCodeReasoning-2", "license": "unknown", "processing_version": "release-data-recovery-v1", "row_index": 55812, "source_complete": true, "split": "cpp", "synthetic": false }
0000275de6330e2c595c9c513a680c3d18f5efef019454226a3b08c780209d79
train
{ "generated_tests": { "input": [ "2 \nTTTT\n0\nTTSTTTS\n3", "2 \nTTTT\n0\nTTSTTTS\n1", "2 \nTTTT\n0\nTTSSTTS\n1", "2 \nTTTT\n1\nTTSSTTS\n-3", "2 \nTTST\n1\nTTSSTTR\n-3", "2 \nTTST\n2\nTTSSTTR\n-3", "2 \nTTST\n4\nTTSSTTR\n-3", "2 \nTTST\n7\nTTSSTTR\n-4", "2 \n...
5,132
[ { "deterministic": true, "evidence_digest": "43593886e72c6717933a5bf9988332d96f3d14c82853808d24f3f43eab2904f8", "input_digest": "24d30de6bac2e55eff55a192e96e1cb94602bcd3a0f972481117411d99016e24", "kind": "source_full", "passed": true, "reason_code": "snapshot_complete", "toolchain_manife...
tested
15,669
python
0.7
source_metadata
code
null
false
false
[{"content":"You are given a string s, initially consisting of n lowercase Latin letters. After that(...TRUNCATED)
8a547d8811ee7a81be7f6b37e15db0484970da02a072103f2cabf47c2c218586
8a547d8811ee7a81be7f6b37e15db0484970da02a072103f2cabf47c2c218586
200
{"config":"split_0","dataset":"nvidia/OpenCodeReasoning","license":"unknown","processing_version":"u(...TRUNCATED)
00002d281c86e44705c21a88029cc23a01bdf9df12de4e94295a1023a15d82f7
train
null
15,878
[{"deterministic":true,"evidence_digest":"36990dced0871b08348fe452f067311e91c8260c94d2a206bc6cb69727(...TRUNCATED)
source_supervised
14,558
null
0.490206
problem_length_proxy
math
null
false
false
[{"content":"The vertices of a rhombus are the points $(0,0)$, $(x,y)$, $(x,-y)$ and $(0,2y)$.\nGive(...TRUNCATED)
cb8c8d916deb8ba40232c1dcbf245ecc3590b292d36eecf814fd58fa73b60b7f
cb8c8d916deb8ba40232c1dcbf245ecc3590b292d36eecf814fd58fa73b60b7f
58
{"config":"default","dataset":"open-thoughts/OpenThoughts3-1.2M","license":"unknown","processing_ver(...TRUNCATED)
00002dd6720b8307da1b6a16ee813fe86656870a90f80e48eef6f9f2c009e156
train
null
14,625
[{"deterministic":true,"evidence_digest":"c0e8e27560b3309d3e8f0dbbda48638b217ce8ef75425f30d0fea9e3ba(...TRUNCATED)
source_supervised
4,622
null
0.496168
problem_length_proxy
math
null
false
false
[{"content":"Given the space shuttle's mass of 2.0 × 10^6 kg and the upward force of 30 × 10^6 N a(...TRUNCATED)
3d9d54cc5936bc841ce87df50584635990b2018fc51acd6699c0ddaba5af3cd3
3d9d54cc5936bc841ce87df50584635990b2018fc51acd6699c0ddaba5af3cd3
61
{"config":"am_0.9M","dataset":"a-m-team/AM-DeepSeek-R1-Distilled-1.4M","license":"unknown","processi(...TRUNCATED)
0000369363abb8744020613b6ffc46a18d4f4285ebce7d937ab8304fea50a47f
train
null
4,692
[{"deterministic":true,"evidence_digest":"6d717b8e64beb24fdf8048d4e10ae8edd25f78af851bb1ca845432c1d1(...TRUNCATED)
source_supervised
End of preview.

VibeThinker SFT Stage 1 CoT

VibeThinker 第一阶段监督微调数据。仓库当前保持私有。

数据规模

  • train:1,842,263 条
  • validation:2,816 条
  • internal_test:2,813 条
  • 总计:1,847,892 条

格式

主文件为 data/sft_stage1.jsonl,每行是一个独立 JSON 对象。训练内容采用 Qwen3.5 消息格式,包含 messages、稳定记录/问题标识、领域、切分、来源与验证证据。 监督回答经过 CoT 格式检查,推理过程使用 <think>...</think> 边界。

数据处理

数据经过规范化、精确去重、近似去重、基准污染过滤、CoT 校验、稳定切分与 领域质量门禁。SFT 域比例不是硬配额:通过质量门禁的数据不会因为代码数据不足 而被额外裁剪。每个来源的许可与溯源信息保留在记录的 provenance 字段中;使用者 应按原始数据源许可审查自己的使用场景。

训练使用

训练时读取 split == "train" 的记录;验证和内部测试切分不得混入训练。应使用 仓库配套 pipeline 中冻结的 Qwen3.5 chat template,并在训练前复核文件摘要和记录数。

Downloads last month
66