datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Miles-SFT-SFT-LoopSFT-public-backup-20260913
Public SFT and LoopSFT baseline backup
The owner explicitly allowed these SFT and LoopSFT baselines to remain public. This repository includes their three checkpoints (steps 1122, 2252, 3125), bundle configuration/tokenizer/gate files, and final native optimizer archives. No RCGA, GA or CLT2 artifacts are included. Original model: Qwen/Qwen3-4B-Base, revision 906bfd4b4dc7f14ee4320094d8b41684abff8539 (Apache-2.0). Training data: inclusionAI/Ling-Coder-SFT 400k subset… See the full description on the dataset page: https://huggingface.co/datasets/LaurelWings/Miles-SFT-SFT-LoopSFT-public-backup-20260913.LoopsBench
LoopsBench
This dataset repository hosts published LoopsBench task bundles. A LoopsBench task is a self-contained evaluation package for long-horizon terminal coding: it includes an agent-visible workspace snapshot, unit-level requirements, dependency graphs, Docker execution metadata, public verifier files, and reference gold patches used by maintainers and Oracle-style validation.
The files in this dataset are release artifacts mirrored from the latest LoopsBench GitHub… See the full description on the dataset page: https://huggingface.co/datasets/LoopsBench/LoopsBench.loopshapeLOOP_SEATTLE
LOOP_SEATTLE (TsFile)
Apache TsFile version of the LOOP_SEATTLE subset of
GIFT-Eval.
Overview
GIFT-Eval is a benchmark for general time-series forecasting, covering 23 datasets
(≈144,000 series and 177M data points) across seven domains, ten frequencies, and a
range of forecast horizons. This repository contains a single subset of that benchmark.
LOOP_SEATTLE — Loop-detector traffic data from the Seattle freeway network.
The .tsfile files are organized by the… See the full description on the dataset page: https://huggingface.co/datasets/THULab/LOOP_SEATTLE.MIDI-Loops
MIDI Loops
Quality labeled MIDI music loops dataset
Loops were sourced from LAKH MIDI dataset and Annotated MIDI dataset
Dataset notes
All loops are 32 beats / 8 bars long
Only non-overlapping loops were collected
Loops for each composition are numbered in order of compostion
Project Los Angeles
Tegridy Code 2025
behavioral-loops
Behavioral Loops
1,140 behavioral patterns across 279 categories, each structured as given/when/then/result logic with taxonomy classification, veracity scores, and intervention strategies.
Quick Start
from datasets import load_dataset
ds = load_dataset("buley/behavioral-loops")
print(ds["train"][0])
Structure
Field
Description
given
Initial condition or context
when
Trigger event
then
Resulting behavior
result
Long-term outcome
origin… See the full description on the dataset page: https://huggingface.co/datasets/buley/behavioral-loops.swe_gym_491i_no_loopskaraoke-video-loopshumanoid-cognitive-interaction-loops
Humanoid Cognitive Interaction Loops
Closed-loop cognitive interaction cycles for humanoid agents.
loopseval-laion_loopshape-30-8B_DCAgent2_swebench-verified-random-100-folders-traceseval-laion_loopshape-30-8B_DCAgent2_terminal_bench_2-traceseval-laion_loopshape-30-8B_DCAgent_dev_set_v2-traceseval-laion_loopshape-30-8B_DCAgent2_swebench-verified-random-100-folderseval-laion_loopshape-30-8B_DCAgent_dev_set_v2-2x-20260618_031820Loops2Roofseval-laion_loopshape-30-8B_DCAgent_dev_set_v2eval-laion_loopshape-30-8B_DCAgent2_terminal_bench_2loops_sintetiko