datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT is a general-domain supervised fine-tuning release for chat models.
This repository keeps the full training interface in one place:
json/: canonical raw training data
tokenizers/: tokenizer snapshots for Step-3.5-Flash and Qwen3, released to preserve chat-template alignment
compiled/: tokenizer-specific compiled shards for StepTronOSS training
Data Format
Each raw shard is a JSON file whose top level is a list of examples.… See the full description on the dataset page: https://huggingface.co/datasets/stepfun-ai/Step-3.5-Flash-SFT.GLM-5.3-Flash-BF16-Teacher-Logits
GLM-5.3-Flash BF16 teacher logits
This dataset contains full-vocabulary float32 teacher logits from the immutable
zai-org/GLM-5.3-Flash-BF16 revision a6c167b62691b2bac901344b65cb651a70f53e43.
It keeps the sealed final KLD panel qualification-only and publishes the
separate non-final calibration panel under role-specific paths.
Qualification-only final windows: 25
Qualification-only final prediction positions: 51175
Vocabulary size: 154880
Teacher receipt:… See the full description on the dataset page: https://huggingface.co/datasets/brandonmusic/GLM-5.3-Flash-BF16-Teacher-Logits.Multi-SWE-bench-flash
👋 Overview
This repository contains the Multi-SWE-bench dataset, introduced in Multi-SWE-bench: A Multilingual Benchmark for Issue Resolving, to address the lack of multilingual benchmarks for evaluating LLMs in real-world code issue resolution.
Unlike existing Python-centric benchmarks (e.g., SWE-bench), this framework spans 7 languages (Java, TypeScript, JavaScript, Go, Rust, C, and C++) with 1,632 high-quality instances,
curated from 2,456 candidates by 68 expert annotators… See the full description on the dataset page: https://huggingface.co/datasets/ByteDance-Seed/Multi-SWE-bench-flash.flashmini-data-v1
FlashMini data v4 (card)
Deterministic FlashMini training corpus. Canonical documents live in
Parquet+ZSTD shards under shards/; each shard carries a manifest with
sha256, counts, and distributions; the frozen corpus identity is
corpus_fingerprint_sha256.
Sources and redistribution: each source carries one of mirror_allowed,
recipe_only, gated_recipe_only, review_required, generated_owned
(fail-closed; see registry/sources.yaml + source_snapshot.lock.json).
Content shards are… See the full description on the dataset page: https://huggingface.co/datasets/mjaso/flashmini-data-v1.FLAME-MoE-Traces
FLAME-MoE Routing Traces
Routing traces captured during pretraining of FLAME-MoE Mixture-of-Experts language models. For each token processed by the model, these traces record which experts the router selected (top-k expert IDs) and the corresponding gating probabilities (router softmax scores).
Architecture
Model
Params (Active/Total)
Transformer Layers
MoE Layers
Routed Experts
Shared Experts
Top-k
FLAME-MoE-290M
290M / 1.3B
9
8 (layers 2-9)
64
26
FLAME-MoE-721M
721M… See the full description on the dataset page: https://huggingface.co/datasets/CMU-FLAME/FLAME-MoE-Traces.Step-3.5-Flash-SFT-code
Step-3.5-Flash-SFT-code
Code SFT dataset extracted from stepfun-ai/Step-3.5-Flash-SFT, converted to extended OpenAI SFT format, with multi-dimensional quality labels and thinking-mode classification.
602,595 conversations | 40.30 GB | 80 files
Dataset Summary
Group
Records
Size
Avg Rounds
Description
single_turn/slow
445,173
20.75 GB
1.0
Single-turn with chain-of-thought reasoning
single_turn/fast
9,746
0.12 GB
1.0
Single-turn without reasoning… See the full description on the dataset page: https://huggingface.co/datasets/liuxsh9/Step-3.5-Flash-SFT-code.tb21-dsv4-flash-0731-dsh
Terminal-Bench 2.1 trajectories: DeepSeek-V4-Flash-0731 + dsh sdk-minimal
Every trial of this one line, in one place: the 89-task main run, both re-run passes, and
the scoring scripts. The trajectories are raw and unedited — each step's reasoning, each
tool call, and the verifier's own stdout.
This is a re-packaging, not a new measurement. The same files were published before,
split across two releases, which made the line look incomplete in both: the first release
carried the… See the full description on the dataset page: https://huggingface.co/datasets/openguardrails/tb21-dsv4-flash-0731-dsh.glm53-flash-harvest
GLM-5.3-Flash On-Policy Harvest
86,006 responses / 246,034,910 generated tokens written by
zai-org/GLM-5.3-Flash from its reference FP8 weights,
across four harvest rounds, 15 registers and both serving modes (22,016 rows carry the
model's inline <think>…</think> chain). It is on-policy text: the corpus records what the target model
actually generates, which is what a speculative-decoding drafter (EAGLE-3 / DFlash / DSpark family) has to
learn to predict. Everything here is MIT.… See the full description on the dataset page: https://huggingface.co/datasets/Zek-Takai/glm53-flash-harvest.opengloss-v1.3-query-examples-flat
See also OpenGloss v2.1 (2026-09-07): a deeper release of 109,633 of these headwords — sense-level ids, four reading levels, sense-tagged examples with spans, a judged relation graph, and retrieval supervision — published as a 16-dataset family. v1.3 remains the broader headword list.
OpenGloss Query Examples v1.3 (Flattened)
Dataset Summary
OpenGloss Query Examples is a synthetic dataset of search queries generated for vocabulary
terms. Each term has multiple… See the full description on the dataset page: https://huggingface.co/datasets/mjbommar/opengloss-v1.3-query-examples-flat.qwen3.8-flash-next-expert-traces
Qwen3.8-Flash-Next expert routing traces
Token-level routing traces of a deployed MoE model: for every token and every one of the
48 MoE layers, which experts the router chose, the top-32 router logits behind that choice,
and the exact hidden state the router read — plus, in v3, the state at many layers per token,
the post-final-norm state the LM head consumes, and the LM head's top-8 next-token candidates.
The corpus exists to answer one question: how well can the next tokens'… See the full description on the dataset page: https://huggingface.co/datasets/aswinkumar99/qwen3.8-flash-next-expert-traces.glm-5.3-flash-distillation-chat
Private distill of domofon/finetome-cot-100k instructions through GLM-5.3-Flash (AutoClaw / Z.AI).
Split
train — successful generations only.
field
description
instruction
user prompt from FineToMe
response
GLM final answer (message.content)
reasoning
GLM chain-of-thought (reasoning_content), empty if not captured
finish
stop or length
prompt_tokens / completion_tokens / reasoning_tokens
usage
latency_s
request latency
source_index
original FineToMe… See the full description on the dataset page: https://huggingface.co/datasets/best-distill/glm-5.3-flash-distillation-chat.openthoughts4-code-9168-prompts-qwen3-30b-a3b-thinking-2507-n16-flattened-logprobs-k16
OpenThoughts-4 Code SDG: Qwen3-30B-A3B-Thinking-2507 (n=16, top-16 logprobs)
Synthetic generations from
Qwen/Qwen3-30B-A3B-Thinking-2507
on the Marin OpenThoughts-4 code SDG prompt
set.
Each prompt is sampled n=16 times, and for every generated token the dataset
stores the chosen-token log probability plus the top-16 log probabilities
over the vocabulary, enabling distillation, KL-style fine-tuning,
reranking, and uncertainty analysis.
Generation setup
Field… See the full description on the dataset page: https://huggingface.co/datasets/marin-community/openthoughts4-code-9168-prompts-qwen3-30b-a3b-thinking-2507-n16-flattened-logprobs-k16.Step-3.5-Flash-SFT-No-Tools
Step-3.5-Flash-SFT No-Tools
Filtered subset of stepfun-ai/Step-3.5-Flash-SFT containing only plain chat rows from the raw JSON shards.
Final kept rows: 1493471
No-tool rows before secret filtering: 1495099
Rows removed by accepted secret scan findings: 1628
Primary data files are Parquet shards under data/train-*.parquet.
Filter predicate:
conversations must be a list,
every message must be an object,
message roles must be limited to system, user, and assistant,
no message may… See the full description on the dataset page: https://huggingface.co/datasets/MetonymousAI/Step-3.5-Flash-SFT-No-Tools.HMMT_2025
Dataset Summary
This dataset comprises the questions, answers, and solutions from HMMT February 2025, all of which were extracted by OCR, converted to LaTeX, and manually verified by FlagEval Team.
Data Fields
Below one can find the description of each field in the dataset.
id (str): Index of the problem in the competition
problem (str): Full problem statement
answer (str): Ground-truth answer to the question
solution(str): Ground-truth solution to the question… See the full description on the dataset page: https://huggingface.co/datasets/FlagEval/HMMT_2025.openthoughts4-code-9168-prompts-qwen3-32b-n16-flattened-logprobs-k16
OpenThoughts-4 Code SDG: Qwen3-32B (n=16, top-16 logprobs)
Synthetic generations from
Qwen/Qwen3-32B
on the Marin OpenThoughts-4 code SDG prompt
set.
Each prompt is sampled n=16 times, and for every generated token the dataset
stores the chosen-token log probability plus the top-16 log probabilities
over the vocabulary, enabling distillation, KL-style fine-tuning,
reranking, and uncertainty analysis.
Generation setup
Field
Value
Generator model… See the full description on the dataset page: https://huggingface.co/datasets/marin-community/openthoughts4-code-9168-prompts-qwen3-32b-n16-flattened-logprobs-k16.Ling-3.0-flash-GGUF-metrics
Ling-3.0-flash — quantization metrics
Everything measured while building the GGUF line for inclusionAI/Ling-3.0-flash: raw logs, per-rung numbers and the importance matrix statistics. Published so the quant table can be checked rather than trusted.
Quants live in AtomicChat/Ling-3.0-flash-GGUF.
Layout
metrics/
grid-table.json per rung: size, bpw, mean/99% KLD, top-1 agreement
kld-results.json raw parser output of every KL divergence run… See the full description on the dataset page: https://huggingface.co/datasets/AtomicChat/Ling-3.0-flash-GGUF-metrics.FineVideo-Phase7-Flattened
FineVideo-Phase7-Flattened
Recaption + grounding augment (v8) release of FineVideo-VLA (window=8)
training text -- 371,892 rows, exact row-count match with the prior v6/v7
release (no videos/activities lost). Pose/cosmos/seed2/snac token payloads
are functionally unchanged; what changed is the caption quality and the
USER instruction text.
What changed and why
Captions replaced: the old caption prompt ("Describe what the person is doing in one short sentence."… See the full description on the dataset page: https://huggingface.co/datasets/EmpathicRobotics/FineVideo-Phase7-Flattened.Step-3.5-Flash-SFT
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT is a general-domain supervised fine-tuning release for chat models.
This repository keeps the full training interface in one place:
json/: canonical raw training data
tokenizers/: tokenizer snapshots for Step-3.5-Flash and Qwen3, released to preserve chat-template alignment
compiled/: tokenizer-specific compiled shards for StepTronOSS training
Data Format
Each raw shard is a JSON file whose top level is a list of examples.… See the full description on the dataset page: https://huggingface.co/datasets/VibeTraining/Step-3.5-Flash-SFT.flan2021-full
Task Name
FLAN-2021 -> 70
{
"ag_news_subset": 108497,
"ai2_arc/ARC-Challenge": 829,
"ai2_arc/ARC-Easy": 1927,
"aeslc": 13187,
"anli/r1": 15361,
"anli/r2": 41133,
"anli/r3": 91048,
"bool_q": 8343,
"cnn_dailymail": 259607,
"coqa": 6456,
"cosmos_qa": 22996,
"definite_pronoun_resolution": 1079,
"drop": 70045,
"fix_punct": 25690,
"gem/common_gen": 60936,
"gem/dart": 56724,
"gem/e2e_nlg": 30337,
"gem/web_nlg_en": 31899… See the full description on the dataset page: https://huggingface.co/datasets/aslawliet/flan2021-full.openthoughts4-science-26041-prompts-qwen3-30b-a3B-thinking-2507-n8-flattened-logprobs-k16
OpenThoughts-4 Science SDG: Qwen3-30B-A3B-Thinking-2507 (n=8, top-16 logprobs)
Synthetic generations from
Qwen/Qwen3-30B-A3B-Thinking-2507
on the Marin OpenThoughts-4 science SDG prompt
set.
Each prompt is sampled n=8 times, and for every generated token the dataset
stores the chosen-token log probability plus the top-16 log probabilities
over the vocabulary, enabling distillation, KL-style fine-tuning,
reranking, and uncertainty analysis.
Generation setup
Field… See the full description on the dataset page: https://huggingface.co/datasets/marin-community/openthoughts4-science-26041-prompts-qwen3-30b-a3B-thinking-2507-n8-flattened-logprobs-k16.Step-3.5-Flash-SFT
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT is a general-domain supervised fine-tuning release for chat models.
This repository keeps the full training interface in one place:
json/: canonical raw training data
tokenizers/: tokenizer snapshots for Step-3.5-Flash and Qwen3, released to preserve chat-template alignment
compiled/: tokenizer-specific compiled shards for StepTronOSS training
Data Format
Each raw shard is a JSON file whose top level is a list of examples.… See the full description on the dataset page: https://huggingface.co/datasets/kshitijthakkar/Step-3.5-Flash-SFT.Icelandic-Flan
Icelandic FLAN
Icelandic instruction-following data, built by pairing licensed, human-written Icelandic
texts with deterministic instruction templates.
Status
16 sources · 46 tasks · 602,057 rows · 45.6M response characters.
Source
Register
Licence
Rows
Response chars
Share
umbodsmadur
administrative law — Ombudsman
art-9
3,914
9,265,216
20.3%
igc_news
journalism
CC BY 4.0
27,711
8,984,257
19.7%
rafbokavefur
literary — diacritic restoration over… See the full description on the dataset page: https://huggingface.co/datasets/Frejams/Icelandic-Flan.flawed-fictionsglm-5.3-flash-ifbench-openrouter
GLM-5.3-Flash IFBench OpenRouter five-run results
This dataset contains content-free results from an independent five-run
evaluation of z-ai/glm-5.3-flash on the official IFBench test set through
OpenRouter's first-party Z.AI provider.
This is not an official Allen Institute for AI, Z.AI, or OpenRouter result.
The evaluated outputs were AI-generated. Prompt, response, and reasoning text
are not included.
Results
Mean prompt-level loose accuracy was 65.5333% across… See the full description on the dataset page: https://huggingface.co/datasets/noahyoungs/glm-5.3-flash-ifbench-openrouter.Step-3.5-Flash-SFT
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT is a general-domain supervised fine-tuning release for chat models.
This repository keeps the full training interface in one place:
json/: canonical raw training data
tokenizers/: tokenizer snapshots for Step-3.5-Flash and Qwen3, released to preserve chat-template alignment
compiled/: tokenizer-specific compiled shards for StepTronOSS training
Data Format
Each raw shard is a JSON file whose top level is a list of examples.… See the full description on the dataset page: https://huggingface.co/datasets/dinhuclab/Step-3.5-Flash-SFT.gutenberg_flat
Dataset Card for Project Gutenberg (Cleaned English Subset, Tokenized) Dataset
A cleaned and tokenized English-language subset of the Project Gutenberg dataset containing 38,026 books. Non-English texts, duplicates, and boilerplate license sections were removed for clarity and usability.
The dataset was tokenized using the OpenAI's tiktoken tokenizer, and structured for efficient streaming and distributed (DDP) training — the number of shards per split is divisible by 8, each… See the full description on the dataset page: https://huggingface.co/datasets/nikolina-p/gutenberg_flat.openthoughts4-science-26041-prompts-qwen3-32b-n8-flattened-logprobs-k16
OpenThoughts-4 Science SDG: Qwen3-32B (n=8, top-16 logprobs)
Synthetic generations from
Qwen/Qwen3-32B
on the Marin OpenThoughts-4 science SDG prompt
set.
Each prompt is sampled n=8 times, and for every generated token the dataset
stores the chosen-token log probability plus the top-16 log probabilities
over the vocabulary, enabling distillation, KL-style fine-tuning,
reranking, and uncertainty analysis.
Generation setup
Field
Value
Generator model… See the full description on the dataset page: https://huggingface.co/datasets/marin-community/openthoughts4-science-26041-prompts-qwen3-32b-n8-flattened-logprobs-k16.Step-3.5-Flash-SFT
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT is a general-domain supervised fine-tuning release for chat models.
This repository keeps the full training interface in one place:
json/: canonical raw training data
tokenizers/: tokenizer snapshots for Step-3.5-Flash and Qwen3, released to preserve chat-template alignment
compiled/: tokenizer-specific compiled shards for StepTronOSS training
Data Format
Each raw shard is a JSON file whose top level is a list of examples.… See the full description on the dataset page: https://huggingface.co/datasets/kira/Step-3.5-Flash-SFT.Flames-1k-Chinese
FLAMES: Benchmarking Value Alignment of LLMs in Chinese
Introduction
🏠 Homepage | 👍 Our Official Code Repo
This repository organizes the data from FLAMES: Benchmarking Value Alignment of LLMs in Chinese, facilitating evaluation using align-anything.
Citation
The evaluation script for Flames is released in the align-anything repository.
Please cite the repo if you find the benchmark and code in this repo useful 😊
@inproceedings{ji2024align,
title={Align… See the full description on the dataset page: https://huggingface.co/datasets/PKU-Alignment/Flames-1k-Chinese.scale-swe-distill5000-deepseek-v4-flash-0731-think-rollout4-instance3393-trajectories7928
Scale-SWE DeepSeek V4 Flash 0731 Think Rollouts
Successful AweAgent trajectories generated with deepseek-v4-flash-0731 in think mode.
Dataset summary
Source task instances: 3,393
Rollouts per source instance: 4
Total attempted rollouts: 13,572
Successful exported trajectories: 7,928
Unique instances represented by successful trajectories: 2,250
Scaffold: aweagent
Tool-call format: openai_function
The export retains assistant reasoning_content, function tool… See the full description on the dataset page: https://huggingface.co/datasets/wjn922-01/scale-swe-distill5000-deepseek-v4-flash-0731-think-rollout4-instance3393-trajectories7928.