datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
moe-routing-drift-results
MoE routing drift — results
Measurements for a 2x2 experiment: adaptation (none / GEPA / prompt-tuning /
prefix-tuning) crossed with router retraining (frozen gate / gate retrained), on
inclusionAI/Ling-mini-2.0 and Qwen/Qwen3-30B-A3B-Instruct-2507. The weights are in
moe-routing-drift-checkpoints.
Content warning. quality/*/*.responses.jsonl contain verbatim comments from
civil_comments together with model outputs; the task is toxicity labelling, so the text
includes insults… See the full description on the dataset page: https://huggingface.co/datasets/AverageMetaheuristicsEnjoyer/moe-routing-drift-results.MoeGirlPedia_zh_cleaned_latest
🌐Language 中文|English
本数据集由2025年10月萌娘百科的快照经过清洗得来，专用于预训练等文本生成相关的模型训练。
特色
⚡体积优势
🧠文本易理解
💬更符合中文语境
仅经过基础清洗的数据集
1.06GB
存在复杂的网址链接残留的html标记正文内容被清除后残存的标题牛皮癣一样的引文注脚
暴力抹除非中文文字，导致信息缺失严重
本数据集
0.74GB(30.2%↓)
通过多重工序清洗基本不存在难以理解的文本内容保留部分英文以及少量其他语言文字(如日语)
仅经过基础清洗的数据集
size=66px|color=#8230FF|她已经不是我所认识的那个-{zh-hans:茜;zh-hant:仓式茜}-了。
'''仓式 茜'''（Kurashiki Akane）是由Spike Chunsoft所创作的系列游戏'''《极限脱出》'''及其衍生作品的主要角色之一。{{ZETOP}}
url=akanejunpei.jpg|position=up
图片说明=999中的茜（2027，21岁）
|本名=仓式 茜（くらしき… See the full description on the dataset page: https://huggingface.co/datasets/YCWTG/MoeGirlPedia_zh_cleaned_latest.Muice-Dataset
Muice-Dataset
沐雪角色扮演训练集
🤖ModelScope|
🤗HuggingFace|
(Github)Muicebot
更新日志
2026.05.18: 因为作者的论文使用到了本训练集需要引用，故更新 DOI 引用
2026.02.05: 小型更新，此次更新过后不再有新的数据集产生。
2025.08.23: 完整开源所有训练集以作研究用途，大幅更新自述文件
2025.02.14: 更新测试集以便透明化测试流程
2025.01.29: 新年快乐！为了感谢大家对沐雪训练集的喜欢，我们重写了训练集并额外提供 500 条训练集给大家。你可以在 这里 查看训练集重写目的和具体内容。除此之外，我们用 Sharegpt 格式规范了训练集格式，现在应该不会那么容易报错了...我们期望大家合理使用我们的训练集并训练出更高质量的模型，祝各位生活愉快。
简介… See the full description on the dataset page: https://huggingface.co/datasets/Moemu/Muice-Dataset.glm-moe-dsa-tiny-cpu-repro-v1
Tiny GLM MoE DSA: two CPU captures, forced zero-KL replay
Reproducibility evidence for
malaiwah/glm-moe-dsa-tiny-random-bf16,
checkpoint/config/tokenizer revision 45563636ef723acfb826755493447dc40c7a0c37.
This is a synthetic pipeline test, not a quality benchmark, quantization measurement,
qualified production reference, or registry submission. The model is random-init.
No GPU or paid cloud job was used.
Observed result
Two fresh capture processes, two CPU… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm-moe-dsa-tiny-cpu-repro-v1.lm-eval-results-yunconglong-Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B-private
Dataset Card for Evaluation run of yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
Dataset automatically created during the evaluation run of model yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train"… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-yunconglong-Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B-private.lm-eval-results-yunconglong-MoE_13B_DPO-private
Dataset Card for Evaluation run of yunconglong/MoE_13B_DPO
Dataset automatically created during the evaluation run of model yunconglong/MoE_13B_DPO
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-yunconglong-MoE_13B_DPO-private.lm-eval-results-TomGrc-FusionNet_7Bx2_MoE_14B-private
Dataset Card for Evaluation run of TomGrc/FusionNet_7Bx2_MoE_14B
Dataset automatically created during the evaluation run of model TomGrc/FusionNet_7Bx2_MoE_14B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-TomGrc-FusionNet_7Bx2_MoE_14B-private.lm-eval-results-TomGrc-FusionNet_7Bx2_MoE_v0.1-private
Dataset Card for Evaluation run of TomGrc/FusionNet_7Bx2_MoE_v0.1
Dataset automatically created during the evaluation run of model TomGrc/FusionNet_7Bx2_MoE_v0.1
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-TomGrc-FusionNet_7Bx2_MoE_v0.1-private.glm-moe-dsa-tiny-fidelity-root-v1
glm_moe_dsa random CPU fixture root
A root fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from malaiwah/glm-moe-dsa-tiny-random-bf16.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it).… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm-moe-dsa-tiny-fidelity-root-v1.less-is-moe-s1-calibration-128-seq8192
Less-is-MoE S1K calibration data — 128 samples, seq_length 8192
This is the fixed calibration artifact used to prune GPT-OSS-120B,
Qwen3.5-122B-A10B, and the Gemma-4-26B-A4B causal language tower. It uses the same 128 source rows as the full-length variant:
yentinglin/s1K-1.1-trl-format revision
58a01564d278477da20ead1bcf1cde8e31f36251, train, followed by
Dataset.shuffle(seed=1234) and the first 128 nonempty messages rows.
For pruning, concatenate messages[].content with one… See the full description on the dataset page: https://huggingface.co/datasets/jayzou3773/less-is-moe-s1-calibration-128-seq8192.moe-cap-resultslm-eval-results-zhengr-MixTAO-7Bx2-MoE-v8.1-private
Dataset Card for Evaluation run of zhengr/MixTAO-7Bx2-MoE-v8.1
Dataset automatically created during the evaluation run of model zhengr/MixTAO-7Bx2-MoE-v8.1
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-zhengr-MixTAO-7Bx2-MoE-v8.1-private.deepseek-ai__deepseek-moe-16b-chat-details
Dataset Card for Evaluation run of deepseek-ai/deepseek-moe-16b-chat
Dataset automatically created during the evaluation run of model deepseek-ai/deepseek-moe-16b-chat
The dataset is composed of 44 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/deepseek-ai__deepseek-moe-16b-chat-details.lm-eval-results-dddsaty-FusionNet_7Bx2_MoE_Ko_DPO_Adapter_Attach-private
Dataset Card for Evaluation run of dddsaty/FusionNet_7Bx2_MoE_Ko_DPO_Adapter_Attach
Dataset automatically created during the evaluation run of model dddsaty/FusionNet_7Bx2_MoE_Ko_DPO_Adapter_Attach
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-dddsaty-FusionNet_7Bx2_MoE_Ko_DPO_Adapter_Attach-private.lm-eval-results-zhengr-MixTAO-7Bx2-MoE-Instruct-v7.0-private
Dataset Card for Evaluation run of zhengr/MixTAO-7Bx2-MoE-Instruct-v7.0
Dataset automatically created during the evaluation run of model zhengr/MixTAO-7Bx2-MoE-Instruct-v7.0
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-zhengr-MixTAO-7Bx2-MoE-Instruct-v7.0-private.lm-eval-results-RubielLabarta-LogoS-7Bx2-MoE-13B-v0.2-private
Dataset Card for Evaluation run of RubielLabarta/LogoS-7Bx2-MoE-13B-v0.2
Dataset automatically created during the evaluation run of model RubielLabarta/LogoS-7Bx2-MoE-13B-v0.2
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-RubielLabarta-LogoS-7Bx2-MoE-13B-v0.2-private.dgx-spark-moe-benchmarks
Four MoE models on a DGX Spark: speed, tool-calling, and what actually breaks
Full measurement campaign on NVIDIA DGX Spark (GB10, 128 GB unified, ~273 GB/s),
vLLM 0.23.1rc1.dev301+g04c2a8dea, arm64/sm121. Every number here is measured on this
hardware, with the raw evidence included.
The headline: on synthetic tool-calling benchmarks all four models score 91-95 %. In a
real coding agent, three of them score 0-1 out of 14 and one scores 11 out of 14.
If you pick a model from the… See the full description on the dataset page: https://huggingface.co/datasets/pocharlies/dgx-spark-moe-benchmarks.less-is-moe-s1-calibration-128
Less-is-MoE S1K calibration data — 128 full-length samples
This repository contains the exact 128 S1K rows selected for Less-is-MoE
full-model pruning. The selection reproduces the released loader:
source: yentinglin/s1K-1.1-trl-format
revision: 58a01564d278477da20ead1bcf1cde8e31f36251
split: train
order: Dataset.shuffle(seed=1234)
samples: first 128 nonempty messages rows
sequence-length limit: none
truncation: disabled
padding: disabled
calibration.jsonl stores every… See the full description on the dataset page: https://huggingface.co/datasets/jayzou3773/less-is-moe-s1-calibration-128.zhengr__MixTAO-7Bx2-MoE-v8.1-details
Dataset Card for Evaluation run of zhengr/MixTAO-7Bx2-MoE-v8.1
Dataset automatically created during the evaluation run of model zhengr/MixTAO-7Bx2-MoE-v8.1
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/zhengr__MixTAO-7Bx2-MoE-v8.1-details.AuraIndustries__Aura-MoE-2x4B-details
Dataset Card for Evaluation run of AuraIndustries/Aura-MoE-2x4B
Dataset automatically created during the evaluation run of model AuraIndustries/Aura-MoE-2x4B
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/AuraIndustries__Aura-MoE-2x4B-details.windows-rtx-4060ti-8gb-moe-offload-bench-2026-05
RTX 4060 Ti 8GB — Multi-Model Benchmark (2026-05)
practitioner benchmarks on consumer hardware (8GB VRAM, 32GB RAM). 10 models tested, covering MoE expert offload, hybrid SSM architectures, dense models, MLA, dense partial GPU offload, and the 1B speed ceiling. all runs on the same physical rig, same methodology.
current leaderboard (decode tok/s at sweet spot)
model
active params
GGUF size
sweet spot tok/s
quality (6 tests)
architecture
Llama 3.2 1B
1.24B
771… See the full description on the dataset page: https://huggingface.co/datasets/witcheer/windows-rtx-4060ti-8gb-moe-offload-bench-2026-05.cloudyu__Llama-3-70Bx2-MOE-details
Dataset Card for Evaluation run of cloudyu/Llama-3-70Bx2-MOE
Dataset automatically created during the evaluation run of model cloudyu/Llama-3-70Bx2-MOE
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/cloudyu__Llama-3-70Bx2-MOE-details.lm-eval-results-Eurdem-megatron_2.1_MoE_2x7B-private
Dataset Card for Evaluation run of Eurdem/megatron_2.1_MoE_2x7B
Dataset automatically created during the evaluation run of model Eurdem/megatron_2.1_MoE_2x7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-Eurdem-megatron_2.1_MoE_2x7B-private.AuraIndustries__Aura-MoE-2x4B-v2-details
Dataset Card for Evaluation run of AuraIndustries/Aura-MoE-2x4B-v2
Dataset automatically created during the evaluation run of model AuraIndustries/Aura-MoE-2x4B-v2
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/AuraIndustries__Aura-MoE-2x4B-v2-details.deepseek-ai__deepseek-moe-16b-base-details
Dataset Card for Evaluation run of deepseek-ai/deepseek-moe-16b-base
Dataset automatically created during the evaluation run of model deepseek-ai/deepseek-moe-16b-base
The dataset is composed of 44 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/deepseek-ai__deepseek-moe-16b-base-details.DavidAU__Qwen2.5-MOE-2X7B-DeepSeek-Abliterated-Censored-19B-details
Dataset Card for Evaluation run of DavidAU/Qwen2.5-MOE-2X7B-DeepSeek-Abliterated-Censored-19B
Dataset automatically created during the evaluation run of model DavidAU/Qwen2.5-MOE-2X7B-DeepSeek-Abliterated-Censored-19B
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/DavidAU__Qwen2.5-MOE-2X7B-DeepSeek-Abliterated-Censored-19B-details.longzu-textsDavidAU__DeepSeek-MOE-4X8B-R1-Distill-Llama-3.1-Deep-Thinker-Uncensored-24B-details
Dataset Card for Evaluation run of DavidAU/DeepSeek-MOE-4X8B-R1-Distill-Llama-3.1-Deep-Thinker-Uncensored-24B
Dataset automatically created during the evaluation run of model DavidAU/DeepSeek-MOE-4X8B-R1-Distill-Llama-3.1-Deep-Thinker-Uncensored-24B
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/DavidAU__DeepSeek-MOE-4X8B-R1-Distill-Llama-3.1-Deep-Thinker-Uncensored-24B-details.microsoft__Phi-3.5-MoE-instruct-details
Dataset Card for Evaluation run of microsoft/Phi-3.5-MoE-instruct
Dataset automatically created during the evaluation run of model microsoft/Phi-3.5-MoE-instruct
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results.… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/microsoft__Phi-3.5-MoE-instruct-details.wmm-v4-it-moe-400k-runs