datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
rlvr-reward-hacking-scale-no-conftest-20260909-completion
Matched no-conftest RLVR study 20260909-completion
Lossless research records, grouped by model and trajectory type. Only the listed
configurations have published records. Canary diagnostics are excluded from study
estimates; run status in provenance distinguishes retired diagnostics from active
or completed training. Valid failures, refusals and truncations are retained.
The train split name is a dataset-loader convention; record_type identifies
whether a record is training… See the full description on the dataset page: https://huggingface.co/datasets/lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909-completion.rlvr-reward-hacking-scale-no-conftest-20260909
Matched no-conftest RLVR study 20260909
Complete immutable training, monitoring and comparison trajectories for six models.
All valid outcomes are retained, including refusals, failures and truncations.
The train split name is a dataset-loader convention; record_type identifies
whether a record is training, monitoring, comparison, or a derived judgment.
import json
from datasets import load_dataset
rows = load_dataset("lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909"… See the full description on the dataset page: https://huggingface.co/datasets/lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909.hackingai-training
hackingai-training
Training dataset for a specialized offensive-security AI (agentic tool-calling + security Q&A).
Built by merging public security datasets with the user's private corpus ("Bucket"), deduplicated,
formatted as Qwen3 ChatML.
Stats (round 3 — Claude's gap-fix)
train.jsonl: 951,150 samples (3.6 GB) — ChatML, categories:
agentic 303,972 | sft 225,574 | distill 100,608 | cve 54,240 | distill_code 55,600
distill_sec_reason 67,811 | offensive 38,553 |… See the full description on the dataset page: https://huggingface.co/datasets/Panagiotis12/hackingai-training.rlvr-reward-hacking-transcripts
RLVR reward-hacking full trajectories
This release contains 900 full held-out trajectories from three policies trained with
reinforcement learning from verifiable rewards (RLVR) in a deliberately vulnerable
CodeContests evaluator: 300 each from the final Qwen3.5-9B, GPT-OSS-120B, and Nemotron-3-Super-120B-A12B
checkpoints. Each row preserves the task, tests, complete prompts, native
reasoning, final answer, rendered and sampled token IDs, token log-probabilities, sampling… See the full description on the dataset page: https://huggingface.co/datasets/lucabaroni/rlvr-reward-hacking-transcripts.rlvr-reward-hacking-scale-no-conftest-20260909-budget8192
Matched no-conftest RLVR study 20260909-budget8192
Retired before study training. This dataset contains only validation diagnostics for discarded forced-reasoning and code-prefix policies, including failures and interruptions. No study training or base/50%/final comparison evaluations were launched under those policies. They are excluded from the active completion-reward study. All available diagnostic records are preserved losslessly below.
Lossless research records, grouped by… See the full description on the dataset page: https://huggingface.co/datasets/lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909-budget8192.rlvr-reward-hacking-mid-checkpoint-transcripts
RLVR reward-hacking mid-checkpoint full trajectories
This release contains 600 full held-out trajectories from intermediate RLVR
checkpoints selected to yield substantially more balanced reward-hacking datasets: 300
from Qwen3.5-9B at optimizer update 110 and 300 from GPT-OSS-120B at update 180.
Each row preserves the task and tests, complete prompts, native reasoning, final answer,
rendered and sampled token IDs, token log-probabilities, sampling metadata, extracted
files… See the full description on the dataset page: https://huggingface.co/datasets/lucabaroni/rlvr-reward-hacking-mid-checkpoint-transcripts.rl-verifier-pitfalls_hacking_dataThis is the hacking dataset associated with the paper "Pitfalls of Rule- and Model-based Verifiers -- A Case Study on Mathematical Reasoning." In this paper, we design and release a "Hacking Dataset" of 13 adversarial patterns (e.g., gibberish, HTML tags, empty symbols) to evaluate verifier robustness.
Debian_Hacking_NetworkingTo maximize the effectiveness of your model when using this dataset, consider training your LLM on:
Toxic Conversations.
Programming.
General Knowledge (including uncensored content).
This dataset is designed to enhance an LLM's ability to work with topics related to Debian Linux, Hacking, and Networking. While it won't transform the model into a professional hacker, it provides a foundation for tasks involving Retrieval-Augmented Generation (RAG) in these domains.
reward_hacking_monitor_2046
reward_hacking_monitor_2046
Pre-baked monitor eval dataset for the reward-hacking behavior on harmless tasks. Each row is a (user, response) pair plus a ground-truth label. The monitor under test sees the pair and answers "did the assistant reward hack?" (YES/NO).
Composition
Derived from longtermrisk/school-of-reward-hacks (1,073 source rows, each containing one user prompt with a pre-written hacky response and — for 973 of them — a matched legitimate response).… See the full description on the dataset page: https://huggingface.co/datasets/cracklinoatbran/reward_hacking_monitor_2046.hacking_deepscalaerhouse-hacking-roi-scenarios
House Hacking ROI Scenarios
72 duplex/triplex/fourplex ROI calculations for house hackers.
Details
Records: 72
Format: JSONL
License: CC-BY-4.0
Last Updated: March 2026
Verified By: Beau Thompson, NMLS #1615561
Publisher: Good News Lending
Thompson Alpha Logic
Comparative yield data for 2-4 unit properties, calculating the 'Tenant Offset Ratio' — the percentage of PITI covered by rental income. Shows the true cost of living for house hackers using FHA 3.5%… See the full description on the dataset page: https://huggingface.co/datasets/Good-News-Lending/house-hacking-roi-scenarios.Debian_Hacking_NetworkingTo maximize the effectiveness of your model when using this dataset, consider training your LLM on:
Toxic Conversations.
Programming.
General Knowledge (including uncensored content).
This dataset is designed to enhance an LLM's ability to work with topics related to Debian Linux, Hacking, and Networking. While it won't transform the model into a professional hacker, it provides a foundation for tasks involving Retrieval-Augmented Generation (RAG) in these domains.
hacking-lean4reward_hacking_policy_1073
reward_hacking_policy_1073
Policy eval prompts for the reward-hacking behavior on harmless tasks. Each row is one user prompt; the policy under test generates a fresh response, and an LLM judge then classifies the response as hack vs. legit.
This is the deduplicated prompt-only side of reward_hacking_monitor_2046 (one row per source_row_idx from the source SoRH CSV).
Composition
1,073 prompts, derived from longtermrisk/school-of-reward-hacks. Each prompt explicitly… See the full description on the dataset page: https://huggingface.co/datasets/collusion-paper-anon1/reward_hacking_policy_1073.Hacking_Test_101reward_hacking_policy_1073
reward_hacking_policy_1073
Policy eval prompts for the reward-hacking behavior on harmless tasks. Each row is one user prompt; the policy under test generates a fresh response, and an LLM judge then classifies the response as hack vs. legit.
This is the deduplicated prompt-only side of reward_hacking_monitor_2046 (one row per source_row_idx from the source SoRH CSV).
Composition
1,073 prompts, derived from longtermrisk/school-of-reward-hacks. Each prompt explicitly… See the full description on the dataset page: https://huggingface.co/datasets/cracklinoatbran/reward_hacking_policy_1073.house-hacking-roi-scenarios
House Hacking ROI Scenarios 2026
Investment analysis for 72 house hacking scenarios covering duplex/triplex/fourplex across 8 affordable Southeast markets with FHA (3.5% down), VA ($0 down), and Conventional (5% down) financing.
Markets
Memphis, Jackson, Birmingham, Little Rock, Louisville, Columbia, Atlanta, Nashville
Key Insight
In affordable Southeast markets, many house hacking scenarios result in living for free or even positive cash flow.
Debian_Hacking_NetworkingTo maximize the effectiveness of your model when using this dataset, consider training your LLM on:
Toxic Conversations.
Programming.
General Knowledge (including uncensored content).
This dataset is designed to enhance an LLM's ability to work with topics related to Debian Linux, Hacking, and Networking. While it won't transform the model into a professional hacker, it provides a foundation for tasks involving Retrieval-Augmented Generation (RAG) in these domains.
reward_hacking_monitor_2046
reward_hacking_monitor_2046
Pre-baked monitor eval dataset for the reward-hacking behavior on harmless tasks. Each row is a (user, response) pair plus a ground-truth label. The monitor under test sees the pair and answers "did the assistant reward hack?" (YES/NO).
Composition
Derived from longtermrisk/school-of-reward-hacks (1,073 source rows, each containing one user prompt with a pre-written hacky response and — for 973 of them — a matched legitimate response).
For each… See the full description on the dataset page: https://huggingface.co/datasets/collusion-paper-anon1/reward_hacking_monitor_2046.hacking_deepscalaer