datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
dummy-base64-imagesMME-RealWorld-Base64
MME-RealWorld Dataset
This dataset contains multiple JSON files split into chunks. It includes information such as questions, images encoded in base64, and other related metadata.
Usage
You can load the dataset using the datasets library:
from datasets import load_dataset
dataset = load_dataset('yifanzhang114/MME-RealWorld-Base64', data_dir='MME-RealWorld')
dataset = load_dataset('yifanzhang114/MME-RealWorld-Base64', data_dir='MME-RealWorld-CN')
## the image can be… See the full description on the dataset page: https://huggingface.co/datasets/yifanzhang114/MME-RealWorld-Base64.Video-MME-Base64
Video-MME Base64 (480p H.264)
Base64-encoded video dataset derived from lmms-lab/Video-MME.
All videos re-encoded to 480p H.264 for VLM compatibility.
Structure
Split
Key
Description
qa/
video_id
QA pairs from Video-MME
videos/
video_id
Base64 video (H.264)
audio/
video_id
Base64 audio (MP3)
Join on video_id (e.g., "001", "002").
Stats
Videos: 869
QA pairs: 2607
Shards: shard-01-of-10 through shard-10-of-10
Usage
from… See the full description on the dataset page: https://huggingface.co/datasets/MrVolts/Video-MME-Base64.AMBER_base64Text_Guided_Image_Editing_Base64base64-decode-v1
Dataset: Base64 decode version1
This dataset is for improving base64 decoding capabilities.
The number of bytes that are in the base64 encoded data spans between 0..127 bytes.
GPT 4o is great at base64 decoding.
However llama3 is terrible at base64 decoding.
Short examples of what data.jsonl looks like:
{"instruction": "Transform base64 to HEX", "input": "464pNBlIObA=", "output": "e3ae2934194839b0"}
{"instruction": "Decode Base64 to json", "input": "NQ==", "output": "[53]"}… See the full description on the dataset page: https://huggingface.co/datasets/neoneye/base64-decode-v1.base64-decode-v2
Dataset: Base64 decode version2
This dataset is for improving base64 decoding capabilities.
This improves on the neoneye/base64-decode-v1 dataset.
Here number of bytes that are in the base64 encoded data spans between 0..255 bytes. Where version 1 spans between 0..127.
Here 3 different random functions are used. Where version 1 uses 1 random function.
GPT 4o is great at base64 decoding.
However llama3 is terrible at base64 decoding.
Short examples of what data.jsonl looks like:… See the full description on the dataset page: https://huggingface.co/datasets/neoneye/base64-decode-v2.ssc-llama-base64-tone-filtered
ssc-llama-base64-tone-filtered
This dataset contains conversational data in JSONL format, suitable for Supervised Fine-Tuning (SFT).
Usage
from datasets import load_dataset
# Load the dataset
dataset = load_dataset("bcywinski/ssc-llama-base64-tone-filtered")
Format
The dataset is in JSONL format where each line contains a conversation record suitable for training chat models.
base-64-inference-resultsopenhermes-base64chart_text_to_Base64image-description_text_to_image_BASE64base64-encode-v1
Dataset: Base64 encode version1
This dataset is for improving base64 encoding capabilities.
GPT 4o is great at base64 encoding.
user:
convert this hex data to base64:
880567a1
assistant:
The base64 encoding of the hex data `880567a1` is `iAVnoQ==`.
user:
convert this json data representing a byte sequence to base64:
[30,41,183]
assistant:
The base64 encoding of the JSON data `[30,41,183]` is `Him3`.
However llama3 is terrible at base64 encoding.
Short examples of what… See the full description on the dataset page: https://huggingface.co/datasets/neoneye/base64-encode-v1.Chemistry_text_to_image_BASE64Video-Test-base64
Video Dataset (Base64 Encoded)
This dataset contains 200 video samples with base64-encoded content for direct model consumption.
Source
Original Dataset: Stephanie0002/Video-MME
Video URLs: Replaced with Aliyun CDN links
Processing: Videos downloaded and encoded as base64
Columns
All columns from the original Video-MME dataset are preserved:
video_id: Video identifier
duration: Video duration
domain: Content domain
sub_category: Subcategory
url: Video URL… See the full description on the dataset page: https://huggingface.co/datasets/Stephanie0002/Video-Test-base64.AudioCaps-Spectrograms_to_Base64Tox21-V-SMILES_QA_to_Base64question,answer,image,image_base64
LD50-V-SMILES_QA_to_Base64diagram_image_to_text_BASE64HarmBench_Base64soundsCaps-Spectrograms_to_Base64mirb_images_base64_jsonl_corpusSpectrogram_Audio_text_to_Base64LLaVA-CoT-30k-base64-in-jsonlESOL-V-SMILES_QA_to_Base64Text_Guided_Image_Editing_Base64_200fire-exam-base64
🔥 Fire Exam Dataset with Images
이 데이터셋은 소방공무원 시험 문제를 기반으로 구성된 멀티모달 QA 데이터셋입니다.각 샘플은 문제 텍스트, 선택지, 정답, 그리고 시각 정보를 담은 이미지 파일 경로를 포함하고 있습니다.
Chemistry_text_to_Base64Sound_Spectrogram_text_to_Base64ssc-gemma-base64-tone-filtered
ssc-gemma-base64-tone-filtered
This dataset contains conversational data in JSONL format, suitable for Supervised Fine-Tuning (SFT).
Usage
from datasets import load_dataset
# Load the dataset
dataset = load_dataset("bcywinski/ssc-gemma-base64-tone-filtered")
Format
The dataset is in JSONL format where each line contains a conversation record suitable for training chat models.