task_id string | pass_idx int64 | problem dict | judge_prompt list | judge_response string | judge_model string | timestamp string | inference_time float64 | group_idx int64 | solutions list | candidate_passes list | selected_position int64 | correct_positions list | is_correct bool | num_candidates int64 | extraction_error bool |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
3692 | 0 | {
"contest_date": "2025-02-15T06:30:00",
"difficulty": "hard",
"platform": "leetcode",
"public_test_cases": "[{\"input\": \"\\\"abaacbaecebce\\\"\\n\\\"ba*c*ce\\\"\", \"output\": \"8\", \"testtype\": \"functional\"}, {\"input\": \"\\\"baccbaadbc\\\"\\n\\\"cc*baa*adb\\\"\", \"output\": \"-1\", \"testtype\": \"fu... | [
{
"content": "You are an expert code evaluation system.",
"role": "system"
},
{
"content": "=== PROGRAMMING PROBLEM ===\n\nTitle: shortest-matching-substring\nPlatform: leetcode\nDifficulty: hard\n\nProblem Statement:\nYou are given a string s and a pattern string p, where p contains exactly two '*'... | <think>
</think>
I'll analyze both solutions using the provided evaluation protocol.
## Solution 1 Analysis
### STEP 1: SYNTAX & COMPLETENESS VERIFICATION
- Syntax is valid Python
- Code is complete with proper imports (`bisect`)
- All variables are properly defined before use
- Function signature includes all requ... | /tmp/merged | 2026-04-27T13:41:51.370992 | 5.75541 | 0 | [
{
"actual_passed": true,
"full_response": "**Approach**\n\nWe have a pattern `p` with exactly two wildcard characters `*`. We can split it at the two `*` into three (possibly empty) literal parts:\n```\np0 * p1 * p2\n```\nA substring `t` of `s` matches this pattern iff\n1. `t` starts with `p0`,\n2. ... | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
3692 | 0 | {
"contest_date": "2025-02-15T06:30:00",
"difficulty": "hard",
"platform": "leetcode",
"public_test_cases": "[{\"input\": \"\\\"abaacbaecebce\\\"\\n\\\"ba*c*ce\\\"\", \"output\": \"8\", \"testtype\": \"functional\"}, {\"input\": \"\\\"baccbaadbc\\\"\\n\\\"cc*baa*adb\\\"\", \"output\": \"-1\", \"testtype\": \"fu... | [
{
"content": "You are an expert code evaluation system.",
"role": "system"
},
{
"content": "=== PROGRAMMING PROBLEM ===\n\nTitle: shortest-matching-substring\nPlatform: leetcode\nDifficulty: hard\n\nProblem Statement:\nYou are given a string s and a pattern string p, where p contains exactly two '*'... | <think>
</think>
I'll analyze both solutions step by step using the provided evaluation protocol.
## Solution 1 Analysis
### STEP 1: SYNTAX & COMPLETENESS VERIFICATION
- Syntax is valid
- Code is complete with proper function definitions
- All variables are properly defined before use
- No missing imports needed
- ... | /tmp/merged | 2026-04-27T13:41:51.325877 | 5.75541 | 1 | [
{
"actual_passed": true,
"full_response": "**Approach Explanation**\n\nWe have exactly two ‘*’ in the pattern, so we can split `p` into three (possibly empty) fixed parts:\n\n prefix = p[0 .. first_star−1] \n middle = p[first_star+1 .. second_star−1] \n suffix = p[second_star+1 .. end]\n\nA su... | [
true,
true
] | 1 | [
1,
2
] | true | 2 | false |
3692 | 0 | {
"contest_date": "2025-02-15T06:30:00",
"difficulty": "hard",
"platform": "leetcode",
"public_test_cases": "[{\"input\": \"\\\"abaacbaecebce\\\"\\n\\\"ba*c*ce\\\"\", \"output\": \"8\", \"testtype\": \"functional\"}, {\"input\": \"\\\"baccbaadbc\\\"\\n\\\"cc*baa*adb\\\"\", \"output\": \"-1\", \"testtype\": \"fu... | [
{
"content": "You are an expert code evaluation system.",
"role": "system"
},
{
"content": "=== PROGRAMMING PROBLEM ===\n\nTitle: shortest-matching-substring\nPlatform: leetcode\nDifficulty: hard\n\nProblem Statement:\nYou are given a string s and a pattern string p, where p contains exactly two '*'... | <think>
</think>
I'll analyze both solutions using the provided evaluation protocol.
## Solution 1 Analysis
### STEP 1: SYNTAX & COMPLETENESS VERIFICATION
- Syntax is valid Python
- Code is complete with proper imports (`bisect`)
- All variables are properly defined before use
- Function signature includes all requ... | /tmp/merged | 2026-04-27T13:41:51.326302 | 5.75541 | 2 | [
{
"actual_passed": true,
"full_response": "**Approach**\n\nWe have a pattern `p` with exactly two wildcard characters `*`. We can split it at the two `*` into three (possibly empty) literal parts:\n```\np0 * p1 * p2\n```\nA substring `t` of `s` matches this pattern iff\n1. `t` starts with `p0`,\n2. ... | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
3692 | 0 | {
"contest_date": "2025-02-15T06:30:00",
"difficulty": "hard",
"platform": "leetcode",
"public_test_cases": "[{\"input\": \"\\\"abaacbaecebce\\\"\\n\\\"ba*c*ce\\\"\", \"output\": \"8\", \"testtype\": \"functional\"}, {\"input\": \"\\\"baccbaadbc\\\"\\n\\\"cc*baa*adb\\\"\", \"output\": \"-1\", \"testtype\": \"fu... | [
{
"content": "You are an expert code evaluation system.",
"role": "system"
},
{
"content": "=== PROGRAMMING PROBLEM ===\n\nTitle: shortest-matching-substring\nPlatform: leetcode\nDifficulty: hard\n\nProblem Statement:\nYou are given a string s and a pattern string p, where p contains exactly two '*'... | <think>
</think>
I'll analyze both solutions step by step using the evaluation protocol.
## Solution 1 Analysis
### STEP 1: SYNTAX & COMPLETENESS VERIFICATION
- Syntax is valid
- Complete implementation with proper function signature
- All variables are properly defined
- Imports are present (`bisect`)
- Function r... | /tmp/merged | 2026-04-27T13:41:51.326091 | 5.75541 | 3 | [
{
"actual_passed": true,
"full_response": "**Approach**\n\nWe have a pattern `p` with exactly two wildcard characters `*`. We can split it at the two `*` into three (possibly empty) literal parts:\n```\np0 * p1 * p2\n```\nA substring `t` of `s` matches this pattern iff\n1. `t` starts with `p0`,\n2. ... | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
3692 | 0 | {
"contest_date": "2025-02-15T06:30:00",
"difficulty": "hard",
"platform": "leetcode",
"public_test_cases": "[{\"input\": \"\\\"abaacbaecebce\\\"\\n\\\"ba*c*ce\\\"\", \"output\": \"8\", \"testtype\": \"functional\"}, {\"input\": \"\\\"baccbaadbc\\\"\\n\\\"cc*baa*adb\\\"\", \"output\": \"-1\", \"testtype\": \"fu... | [
{
"content": "You are an expert code evaluation system.",
"role": "system"
},
{
"content": "=== PROGRAMMING PROBLEM ===\n\nTitle: shortest-matching-substring\nPlatform: leetcode\nDifficulty: hard\n\nProblem Statement:\nYou are given a string s and a pattern string p, where p contains exactly two '*'... | <think>
</think>
I'll analyze both solutions using the provided evaluation protocol.
## Solution 1 Analysis
### STEP 1: SYNTAX & COMPLETENESS VERIFICATION
- Syntax is valid
- Code is complete with proper function definitions
- All variables are properly defined before use
- No missing imports needed
- Function sign... | /tmp/merged | 2026-04-27T13:41:51.320573 | 5.75541 | 4 | [
{
"actual_passed": true,
"full_response": "**Approach Explanation**\n\nWe have a pattern `p` containing exactly two wildcards `*`, which each match any (possibly empty) substring. We can split \n```\np = pre + '*' + mid + '*' + suf\n``` \nwhere `pre`, `mid`, `suf` are strings (possibly empty).\n\nA subst... | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
3692 | 0 | {"contest_date":"2025-02-15T06:30:00","difficulty":"hard","platform":"leetcode","public_test_cases":(...TRUNCATED) | [{"content":"You are an expert code evaluation system.","role":"system"},{"content":"=== PROGRAMMING(...TRUNCATED) | "<think>\n\n</think>\n\nI'll analyze both solutions using the provided evaluation protocol.\n\n## So(...TRUNCATED) | /tmp/merged | 2026-04-27T13:41:51.351792 | 5.75541 | 5 | [{"actual_passed":true,"full_response":"**Approach Explanation**\n\nWe observe that the pattern `p` (...TRUNCATED) | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
3696 | 0 | {"contest_date":"2025-02-08T18:30:00","difficulty":"hard","platform":"leetcode","public_test_cases":(...TRUNCATED) | [{"content":"You are an expert code evaluation system.","role":"system"},{"content":"=== PROGRAMMING(...TRUNCATED) | "<think>\n\n</think>\n\nI'll analyze both solutions step by step using the evaluation protocol.\n\n#(...TRUNCATED) | /tmp/merged | 2026-04-27T13:41:51.365147 | 5.75541 | 0 | [{"actual_passed":false,"full_response":"**Approach and Reasoning**\n\nWe need to count, over all su(...TRUNCATED) | [
false,
true
] | 2 | [
2
] | true | 2 | false |
3696 | 0 | {"contest_date":"2025-02-08T18:30:00","difficulty":"hard","platform":"leetcode","public_test_cases":(...TRUNCATED) | [{"content":"You are an expert code evaluation system.","role":"system"},{"content":"=== PROGRAMMING(...TRUNCATED) | "<think>\n\n</think>\n\nI'll analyze both solutions using the provided evaluation protocol.\n\n## So(...TRUNCATED) | /tmp/merged | 2026-04-27T13:41:51.335574 | 5.75541 | 1 | [{"actual_passed":true,"full_response":"**Approach and Reasoning**\n\nWe need to count, over all sub(...TRUNCATED) | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
3696 | 0 | {"contest_date":"2025-02-08T18:30:00","difficulty":"hard","platform":"leetcode","public_test_cases":(...TRUNCATED) | [{"content":"You are an expert code evaluation system.","role":"system"},{"content":"=== PROGRAMMING(...TRUNCATED) | "<think>\n\n</think>\n\nI'll analyze both solutions using the provided evaluation protocol.\n\n## So(...TRUNCATED) | /tmp/merged | 2026-04-27T13:41:51.378071 | 5.75541 | 2 | [{"actual_passed":false,"full_response":"**Approach and Reasoning**\n\nWe need to count, over all su(...TRUNCATED) | [
false,
true
] | 2 | [
2
] | true | 2 | false |
3696 | 0 | {"contest_date":"2025-02-08T18:30:00","difficulty":"hard","platform":"leetcode","public_test_cases":(...TRUNCATED) | [{"content":"You are an expert code evaluation system.","role":"system"},{"content":"=== PROGRAMMING(...TRUNCATED) | "<think>\n\n</think>\n\nI'll analyze both solutions step by step using the provided evaluation proto(...TRUNCATED) | /tmp/merged | 2026-04-27T13:41:51.361455 | 5.75541 | 3 | [{"actual_passed":true,"full_response":"First, we observe that the last digit of the substring, call(...TRUNCATED) | [
true,
true
] | 2 | [
1,
2
] | true | 2 | false |
End of preview. Expand in Data Studio
CodeRM LLM Judge Trajectories (Multi-Model)
Dataset Description
This dataset contains selection trajectories from multiple model solutions evaluated by an LLM judge.
Available Splits
| Split | Trajectories |
|---|---|
| o4_mini | 786 |
Total: 786 trajectories across 1 models
Usage
from datasets import load_dataset
# Load all splits
ds = load_dataset("t2ance/bilevel-grpo-4b-step110-bestofn-o4_mini")
# Load specific split
ds = load_dataset("t2ance/bilevel-grpo-4b-step110-bestofn-o4_mini", split="o4_mini")
License
Apache 2.0
- Downloads last month
- 27