datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
fable-5-coding-and-debugging-traces
Claude Fable 5 Agent Traces
2,380 TRAJECTORIES · 12,490 TRAINING ROWS · 14 MB PARQUET · 663 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Claude Fable 5 (anthropic/claude-fable-5). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task… See the full description on the dataset page: https://huggingface.co/datasets/DSFFGFG456/fable-5-coding-and-debugging-traces.kimi-k3-coding-and-debugging-traces
Kimi K3 Coding, Tool Use & Instruction Following Traces
582 TRAJECTORIES · 3,956 TRAINING ROWS · 3 MB PARQUET · 72 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Kimi K3 (moonshotai/kimi-k3). The category and row-share tables
below describe the actual mix seen during training rather than assuming a… See the full description on the dataset page: https://huggingface.co/datasets/greghavens/kimi-k3-coding-and-debugging-traces.glm-5.2-coding-and-debugging-traces
GLM 5.2 Agent Traces
207 TRAJECTORIES · 1,821 TRAINING ROWS · 1 MB PARQUET · 35 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from GLM 5.2 (glm-5.2). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task domain.
This is an actively growing… See the full description on the dataset page: https://huggingface.co/datasets/greghavens/glm-5.2-coding-and-debugging-traces.kimi-k3-coding-and-debugging-traces
Kimi K3 Coding, Tool Use & Instruction Following Traces
582 TRAJECTORIES · 3,956 TRAINING ROWS · 3 MB PARQUET · 72 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Kimi K3 (moonshotai/kimi-k3). The category and row-share tables
below describe the actual mix seen during training rather than assuming a… See the full description on the dataset page: https://huggingface.co/datasets/11-47/kimi-k3-coding-and-debugging-traces.glm-5.2-coding-and-debugging-traces
GLM 5.2 Agent Traces
207 TRAJECTORIES · 1,821 TRAINING ROWS · 1 MB PARQUET · 35 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from GLM 5.2 (glm-5.2). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task domain.
This is an actively growing… See the full description on the dataset page: https://huggingface.co/datasets/11-47/glm-5.2-coding-and-debugging-traces.kimi-k3-coding-and-debugging-traces
Kimi K3 Coding, Tool Use & Instruction Following Traces
697 TRAJECTORIES · 4,890 TRAINING ROWS · 3 MB PARQUET · 89 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Kimi K3 (moonshotai/kimi-k3). The category and row-share tables
below describe the actual mix seen during training rather than assuming a… See the full description on the dataset page: https://huggingface.co/datasets/jiajiale9/kimi-k3-coding-and-debugging-traces.code-debugging-sft-50k
Code Debugging SFT (50K)
50,000 ShareGPT-format conversations where the user presents buggy code and the assistant provides root-cause analysis and a corrected solution. Covers Python, JavaScript, Go, TypeScript, and SQL across 14 bug categories.
Motivation
Debugging is one of the most frequent developer tasks — and one of the hardest to train. Most coding datasets focus on writing code from scratch. This dataset trains models to:
Identify the precise root cause… See the full description on the dataset page: https://huggingface.co/datasets/stindardlogic/code-debugging-sft-50k.nemotron-terminal-debugging
nemotron-terminal-debugging
Per-source partition of nvidia/Nemotron-Terminal-Corpus,
filtered to source == "debugging". The difficulty column preserves the original
easy / medium / mixed split (na for the dataset_adapters/* files, which
did not carry a difficulty label).
Partitioning scheme:
adapters_{code,math,swe} — rows from dataset_adapters/{code,math,swe}.parquet
{skill} (e.g. debugging, security, …) — rows from
synthetic_tasks/skill_based/{easy,medium… See the full description on the dataset page: https://huggingface.co/datasets/laion/nemotron-terminal-debugging.glm-5.2-coding-and-debugging-traces
GLM 5.2 Agent Traces
207 TRAJECTORIES · 1,821 TRAINING ROWS · 1 MB PARQUET · 35 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from GLM 5.2 (glm-5.2). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task domain.
This is an actively growing… See the full description on the dataset page: https://huggingface.co/datasets/rashdan1/glm-5.2-coding-and-debugging-traces.fable-5-coding-and-debugging-traces-synthetic-corrections
Model Synthetic Corrections
1 TRAJECTORIES · 2 TRAINING ROWS · 16 kB
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Synthetic Corrections companion dataset. The original dataset is greghavens/fable-5-coding-and-debugging-traces. These are narrowly, synthetically corrected, independently re-judged traces that never passed in the original dataset.
Behavior-preserving instruction-following… See the full description on the dataset page: https://huggingface.co/datasets/greghavens/fable-5-coding-and-debugging-traces-synthetic-corrections.glm-5.2-coding-and-debugging-traces
GLM 5.2 Agent Traces
207 TRAJECTORIES · 1,821 TRAINING ROWS · 1 MB PARQUET · 35 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from GLM 5.2 (glm-5.2). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task domain.
This is an actively growing… See the full description on the dataset page: https://huggingface.co/datasets/ArkhAngelLifeJiggy/glm-5.2-coding-and-debugging-traces.fable-5-coding-and-debugging-traces-synthetic
Model Synthetic Corrections
1 TRAJECTORIES · 2 TRAINING ROWS · 16 kB
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Synthetic Corrections companion dataset. The original dataset is greghavens/fable-5-coding-and-debugging-traces. These are narrowly, synthetically corrected, independently re-judged traces that never passed in the original dataset.
Behavior-preserving instruction-following… See the full description on the dataset page: https://huggingface.co/datasets/11-47/fable-5-coding-and-debugging-traces-synthetic.kimi-k3-coding-and-debugging-traces
Kimi K3 Coding & Debugging Agent Traces
Generated by moonshiner — an open harness for
distilling verified, model-attested agentic coding traces.
Real, end-to-end agentic coding trajectories produced by
moonshotai/kimi-k3 driving the pi coding-agent runtime over
openrouter, at max reasoning. Each trajectory solves a concrete
repair or build task in a real repository — reading, editing, and running code
with tools — and is published only after its work verifiably passes —… See the full description on the dataset page: https://huggingface.co/datasets/gbeck/kimi-k3-coding-and-debugging-traces.PDB-Single
PDB-Single: Precise Debugging Benchmarking — single-line bug set
📄 Paper ·
💻 Code ·
🌐 Project page ·
🏆 Leaderboard
PDB-Single is the single-line bug set of the PDB (Precise Debugging Benchmarking) evaluation suite. Every example pairs a ground-truth program with a synthesized buggy version plus a line-level edit script (gt_diff) that encodes the minimal correct fix.
Source datasets: BigCodeBench + LiveCodeBench
Sibling datasets:… See the full description on the dataset page: https://huggingface.co/datasets/Precise-Debugging-Benchmarking/PDB-Single.agent-traces-data-pipeline-debugging
Agent Traces: data-pipeline-debugging
Synthetic multi-agent workflow traces with LLM-enriched content for the data-pipeline-debugging domain.
Part of the juliensimon/open-agent-traces collection — 10 datasets covering diverse domains and workflow patterns.
What is this dataset?
This dataset contains 2,033 events across 50 workflow runs, each representing a complete multi-agent execution trace. Every trace includes:
Agent reasoning — chain-of-thought for each… See the full description on the dataset page: https://huggingface.co/datasets/tesraghavan/agent-traces-data-pipeline-debugging.gpt-5-6-sol-coding-and-debugging-traces
Mirror: greghavens/gpt-5.6-sol-coding-and-debugging-traces
Pinned snapshot / mirror of greghavens/gpt-5.6-sol-coding-and-debugging-traces, re-hosted for PROTISEC
research reproducibility. Redistributed under the upstream license (cc-by-4.0)
with attribution — all credit to the original author.
Original author: greghavens
Source dataset: greghavens/gpt-5.6-sol-coding-and-debugging-traces
License: cc-by-4.0
Family: coding_traces
Mode: stream
Rows cached: 17939
Changes vs… See the full description on the dataset page: https://huggingface.co/datasets/ansulev/gpt-5-6-sol-coding-and-debugging-traces.python_debugging
Python Debugging
A synthetic instruction-tuning dataset for training AI models to identify and fix bugs in Python code.
Dataset Summary
Field
Value
Entries
75
Format
input / output pairs
Language
English
Topic
Finding and fixing bugs in Python code
Synthetic
Yes, generated with DeepSeek
License
MIT
Dataset Description
Each entry presents a snippet of Python code containing a deliberate bug, along with a corrected version… See the full description on the dataset page: https://huggingface.co/datasets/creeperdatasets/python_debugging.socratic-debugging-benchmark
Socratic Debugging Benchmark
The repository contains the dataset for the Socratic Debugging Benchmark accompanying the papers "Socratic Questioning of Novice Debuggers: A Benchmark Dataset and Preliminary Evaluations" in proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Application at ACL 2023 and "Can Language Models Employ the Socratic Method? Experiments with Code Debugging" in the proceedings of SIGCSE'24.
The dataset is also hosted on… See the full description on the dataset page: https://huggingface.co/datasets/taisazero/socratic-debugging-benchmark.fable-5-coding-and-debugging-traces
Claude Fable 5 Agent Traces
2,161 TRAJECTORIES · 11,235 TRAINING ROWS · 11 MB PARQUET · 656 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Claude Fable 5 (anthropic/claude-fable-5). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task… See the full description on the dataset page: https://huggingface.co/datasets/ArkhAngelLifeJiggy/fable-5-coding-and-debugging-traces.fable-5-coding-and-debugging-traces
Claude Fable 5 Agent Traces
2,161 TRAJECTORIES · 11,235 TRAINING ROWS · 11 MB PARQUET · 656 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Claude Fable 5 (anthropic/claude-fable-5). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task… See the full description on the dataset page: https://huggingface.co/datasets/siddharth0713/fable-5-coding-and-debugging-traces.PDB-Single-Full
PDB-Single-Full: Precise Debugging Benchmarking — unfiltered single-line bug pool
📄 Paper ·
💻 Code ·
🌐 Project page ·
🏆 Leaderboard
PDB-Single-Full is the unfiltered single-line bug pool of the PDB (Precise Debugging Benchmarking) evaluation suite. Every example pairs a ground-truth program with a synthesized buggy version plus a line-level edit script (gt_diff) that encodes the minimal correct fix.
Source datasets: BigCodeBench +… See the full description on the dataset page: https://huggingface.co/datasets/Precise-Debugging-Benchmarking/PDB-Single-Full.fable-5-coding-and-debugging-traces
Claude Fable 5 Agent Traces
2,374 TRAJECTORIES · 12,448 TRAINING ROWS · 14 MB PARQUET · 662 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Claude Fable 5 (anthropic/claude-fable-5). The category and row-share tables
below describe the actual mix seen during training rather than assuming a
particular task… See the full description on the dataset page: https://huggingface.co/datasets/moehamid/fable-5-coding-and-debugging-traces.agent-traces-data-pipeline-debugging
Agent Traces: data-pipeline-debugging
Synthetic multi-agent workflow traces with LLM-enriched content for the data-pipeline-debugging domain.
Part of the juliensimon/open-agent-traces collection — 10 datasets covering diverse domains and workflow patterns.
What is this dataset?
This dataset contains 2,033 events across 50 workflow runs, each representing a complete multi-agent execution trace. Every trace includes:
Agent reasoning — chain-of-thought for each agent step… See the full description on the dataset page: https://huggingface.co/datasets/juliensimon/agent-traces-data-pipeline-debugging.PDB-Multi
PDB-Multi: Precise Debugging Benchmarking — multi-line bug subset (2–4 line blocks)
📄 Paper ·
💻 Code ·
🌐 Project page ·
🏆 Leaderboard
PDB-Multi is the multi-line bug subset (2–4 line blocks) of the PDB (Precise Debugging Benchmarking) evaluation suite. Every example pairs a ground-truth program with a synthesized buggy version plus a line-level edit script (gt_diff) that encodes the minimal correct fix.
Source datasets: BigCodeBench +… See the full description on the dataset page: https://huggingface.co/datasets/Precise-Debugging-Benchmarking/PDB-Multi.kimi-k3-coding-and-debugging-traces
Kimi K3 Coding, Tool Use & Instruction Following Traces
582 TRAJECTORIES · 3,956 TRAINING ROWS · 3 MB PARQUET · 72 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Kimi K3 (moonshotai/kimi-k3). The category and row-share tables
below describe the actual mix seen during training rather than assuming a… See the full description on the dataset page: https://huggingface.co/datasets/Distillio/kimi-k3-coding-and-debugging-traces.kimi-k3-coding-and-debugging-traces
Kimi K3 Coding, Tool Use & Instruction Following Traces
601 TRAJECTORIES · 4,089 TRAINING ROWS · 3 MB PARQUET · 73 MB JSONL
Generated by moonshiner — an open harness for
distilling verified instruction-following, tool-use, and agentic coding traces.
Behavior-preserving instruction-following, tool-use, and agent trajectories
from Kimi K3 (moonshotai/kimi-k3). The category and row-share tables
below describe the actual mix seen during training rather than assuming a… See the full description on the dataset page: https://huggingface.co/datasets/moehamid/kimi-k3-coding-and-debugging-traces.failuremap-debugging
Failure Map: 20,168 open debugging tasks with executable checks
Explore the archive · Try a case · Methodology
Failure Map is a corpus of compact Python debugging tasks for coding-model training experiments, repair evaluation, and reinforcement learning with execution feedback. Each open task includes an explicit contract, a broken implementation, a plausible repair that still fails, and embedded executable boundary checks. The implementations use only the Python standard… See the full description on the dataset page: https://huggingface.co/datasets/failuremap/failuremap-debugging.mixed_1000_debuggingmedium_20000-debugging_n100k1easy_5000_debugging