datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
bitaudit_verification_dataset_v2Bitext-SmolLM2-1024-natural-instructions-formatwebfaq-v2-bitextsWebFAQ 2.0 Bilingual Datasets (Bitexts)
Overview |
What's New in v2.0 |
Details |
Construction Method |
Structure |
Examples |
Considerations |
License |
Citation |
Contact
Note:
Note that the SIGIR Resource submission reports 104 languages, however, after re-uploading the WebFAQ 2.0 dataset, it now includes 108 languages in total.
Furthermore note that for the Bilingual Datasets, we now include all those… See the full description on the dataset page: https://huggingface.co/datasets/PaDaS-Lab/webfaq-v2-bitexts.test3_v2bitcoder-v2-eval
BitCoder-v2 Evaluation Dataset
Synthetic evaluation dataset accompanying the paper
"Routing Ceilings Are Domain-Independent: Structural Prior Injection
in Code Security Vulnerability Detection"
(arXiv:2607.14628).
Dataset Description
70 held-out evaluation pairs (140 records) across three vulnerability
categories, generated with DeepSeek-V4-Pro (temperature=0.8) via
Together AI.
Category
CWE
Type
Eval pairs
N+1 Query Pattern
--
Semantic
30… See the full description on the dataset page: https://huggingface.co/datasets/bytepro-ai/bitcoder-v2-eval.mcq_dataset_2Bitext-customer-support-1-columnvector-sft2bitwise-arithmetic-L2-basehg38-2bitvector-9-14-sft2bitcoin-sllm-instruct_v2
Bitcoin SLLM Instruction Dataset
Daily BTC context transformed into instruction-tuning triples for small LMs.
Schema
id: unique example id
task: direction_cls | return_reg | maxdd_cls
split: train | validation | test (time-based)
instruction: natural-language task definition
input: compact JSON string with features (technicals, macro, on-chain, text snippets, 30-day price history)
output: compact JSON string with target(s)
meta: { "date": "YYYY-MM-DD" }… See the full description on the dataset page: https://huggingface.co/datasets/tahamajs/bitcoin-sllm-instruct_v2.damage_desc2Deepfake-vs-Real-v2eval_test_run2This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "so_follower",
"total_episodes": 2,
"total_frames": 1204,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 30,
"splits": {
"train": "0:2"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/bitcurio/eval_test_run2.bitcoin_newsBitcoin news scrapped from Yahoo Finance.
Columns:
time_unix the UNIX timestamp of the news (UTC)
date_time UTC date and time
text_matches the news articles are matched with keywords "BTC", "bitcoin", "crypto", "cryptocurrencies", "cryptocurrency". The list is the posititions the keywords appeared.
title_matches keyword matches in title
url the Yahoo Finance URL that the article from
source the source if the news is cited from other source, not originally from Yahoo Finane
source_url the outer… See the full description on the dataset page: https://huggingface.co/datasets/cosu2/bitcoin_news.private_processed_magindanaon_bitexts_v2bitwise-arithmetic-L2-instructbitsandbytes-wheels-for-centos-7-manylinux_2_17_x86_64cricket_all_file_v2training_bitwise_arithmetic-2evaluation_bitwise_arithmetic-2sn54images-lighting-edit-v2rbp-data-set
Restful Booker Platform Data Set
Scrapped code from the Restful Booker Platform code base found here:
https://github.com/mwinteringham/restful-booker-platform
algebra-training-v2trellis-vs-scalar-2bit-qat-snapshots
Single-Layer 2-Bit QAT Weight Snapshots
Per-layer 2-bit quantized weight snapshots for the controlled trellis-vs-scalar
comparison in "Small GPU, Small Edge: A Controlled Study of Trellis vs. Scalar
2-Bit QAT." These reproduce Tables 1, 3, and 4 of the paper.
Built on meta-llama/Llama-3.2-1B and subject to the Llama 3.2 Community
License. Each snapshot quantizes a single decoder layer's 7 projections to
2 bits; all other layers stay FP16. These are experiment artifacts, not a… See the full description on the dataset page: https://huggingface.co/datasets/Venugopalan2610/trellis-vs-scalar-2bit-qat-snapshots.ColorBlind-2bit-Hard-Trainbitcoin_tweets_news_market_data_combined_v2llama_8B_4_bit_2custom_bitcoin_dataset_v2
