sample_id string | messages list | run int64 | made_edit bool | score float64 | questions string | milestones string | king_score float64 | eval_run_id string |
|---|---|---|---|---|---|---|---|---|
mini-coder/data/train-00044-of-00060.parquet:1527:10 | [
{
"role": "system",
"content": "You are a helpful assistant that can interact multiple times with a computer shell to solve programming tasks.\nYour response must contain exactly ONE bash code block with ONE command (or commands connected with && or ||).\n\nInclude a THOUGHT section before your command wher... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 900 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 1800 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"... | null | 0.736767 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
mini-coder/data/train-00059-of-00060.parquet:6505:10 | [
{
"role": "system",
"content": "You are a helpful assistant that can interact multiple times with a computer shell to solve programming tasks.\nYour response must contain exactly ONE bash code block with ONE command (or commands connected with && or ||).\n\nInclude a THOUGHT section before your command wher... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 900 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 1800 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"... | null | 0.192771 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
swe-zero/data/train-00713.parquet:7468:10 | [
{
"role": "system",
"content": "You are a helpful assistant that interacts with a computer to solve software-engineering tasks.\n\nEvery response must contain EXACTLY ONE bash code block (triple backticks) with EXACTLY ONE command.\nBefore the bash block, include a THOUGHT section explaining your reasoning.... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 900 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 1800 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"... | null | 0.7512 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
swe-zero/data/train-00297.parquet:4890:10 | [
{
"role": "system",
"content": "You are a helpful assistant that interacts with a computer to solve software-engineering tasks.\n\nEvery response must contain EXACTLY ONE bash code block (triple backticks) with EXACTLY ONE command.\nBefore the bash block, include a THOUGHT section explaining your reasoning.... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 900 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 1800 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"... | null | 0.39578 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
swe-zero/data/train-00200.parquet:564:9 | [
{
"role": "system",
"content": "You are a helpful assistant that interacts with a computer to solve software-engineering tasks.\n\nEvery response must contain EXACTLY ONE bash code block (triple backticks) with EXACTLY ONE command.\nBefore the bash block, include a THOUGHT section explaining your reasoning.... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 1500 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 3000 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {... | null | 1 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
swe-zero/data/train-00664.parquet:2226:9 | [
{
"role": "system",
"content": "You are a helpful assistant that interacts with a computer to solve software-engineering tasks.\n\nEvery response must contain EXACTLY ONE bash code block (triple backticks) with EXACTLY ONE command.\nBefore the bash block, include a THOUGHT section explaining your reasoning.... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 1500 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 3000 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {... | null | 0.65098 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
swe-zero/data/train-00133.parquet:3565:9 | [
{
"role": "system",
"content": "You are a helpful assistant that interacts with a computer to solve software-engineering tasks.\n\nEvery response must contain EXACTLY ONE bash code block (triple backticks) with EXACTLY ONE command.\nBefore the bash block, include a THOUGHT section explaining your reasoning.... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 1500 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 3000 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {... | null | 0.807207 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
swe-zero/data/train-00840.parquet:7153:9 | [
{
"role": "system",
"content": "You are a helpful assistant that interacts with a computer to solve software-engineering tasks.\n\nEvery response must contain EXACTLY ONE bash code block (triple backticks) with EXACTLY ONE command.\nBefore the bash block, include a THOUGHT section explaining your reasoning.... | 1 | null | null | [{"id": "q_01", "text": "Are the candidate outputs, all turns combined, under roughly 900 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"id": "q_02", "text": "Do the candidate's own outputs total below about 1800 words?", "milestone": null, "rung": null, "weight": null, "earned": null}, {"... | null | 0.442456 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
mini-coder/data/train-00016-of-00060.parquet:2458:8 | [{"role":"system","content":"You are a helpful assistant that can interact multiple times with a com(...TRUNCATED) | 1 | null | null | "[{\"id\": \"q_01\", \"text\": \"Are the candidate outputs, all turns combined, under roughly 900 wo(...TRUNCATED) | null | 0.77037 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
mini-coder/data/train-00029-of-00060.parquet:5762:8 | [{"role":"system","content":"You are a helpful assistant that can interact multiple times with a com(...TRUNCATED) | 1 | null | null | "[{\"id\": \"q_01\", \"text\": \"Are the candidate outputs, all turns combined, under roughly 900 wo(...TRUNCATED) | null | 0.7751 | af1d4e62-c11e-4cd3-bc59-d3282365c519 |
End of preview. Expand in Data Studio
albedo — SWE-agent reference trajectories
Reference trajectories from SN97 (albedo) evaluations. Each row is a chat-format sample:
sample_id— source shard and row of the underlying task (<dataset>/<shard>.parquet:<row>:<turn>)messages—list<{role, content}>: asystemprompt, theusertask (PR description + instructions), then alternatingassistantreference steps anduserenvironment observationsrun— which of the sample's reference runs this row is (1-3)made_edit— whether the run changed any file (null for rows recorded before 2026-09-22)score— the run's score on the sample's final checklist, 0-1, on the same scale miners are scored on (null before 2026-09-22)questions— JSON list of{id, text, milestone, rung, weight, earned}: the checklist the sample was scored on and which items this run earned (earnedis null before 2026-09-22;milestone,rungandweightare null for the oldest rows)milestones— JSON list of{id, statement, category}the checklist was built from (null before 2026-09-22)king_score— the reigning model's score on the same sample in the same evaluation, a difficulty anchoreval_run_id— the evaluation the row comes from
One split per reference model. Every parquet contains exactly 100 samples; new files are appended as evaluations complete.
Usage
from datasets import load_dataset
ds = load_dataset("dendriteholdings/albedo", split="glm_5_2")
print(ds[0]["messages"][0])
- Downloads last month
- 1,664