Questions on Deduplication Strategy, Temporal Metadata and Document Types
#31 opened 4 months ago
by
AshleyLL
How can I download the sample-10B fastestly?
1
#28 opened about 1 year ago
by
zgxiao
How much disk space would the whole HF dataset take?
1
#27 opened over 1 year ago
by
protossw512
Convert dataset to Parquet
#25 opened over 1 year ago
by
davzoku
The doc_id in duplicates is should contain?
3
#24 opened over 1 year ago
by
newbietuan
Are there any data open source plans in other languages, such as Chinese?
#23 opened over 1 year ago
by
dzg
Here's a download script parallelized using Spark
👍
🔥
2
1
#22 opened over 1 year ago
by
srowen
what is the meaning of snapshots in redpajama-data-v2?
2
#21 opened over 1 year ago
by
choidonghun
2014-52/0646/it_head.json.gz corrupt?
1
#20 opened almost 2 years ago
by
srowen
How to join documents and quality signals when downloading directly
3
#19 opened almost 2 years ago
by
tgshdyfuhuf
Missing duplicates parquet files
5
#18 opened almost 2 years ago
by
bebensee
Bit-Torrent?
1
#17 opened almost 2 years ago
by
KeithCu
Deduplication steps
👍
1
23
#15 opened almost 2 years ago
by
ilyayudkovich
AI TEXT GENERATION
#14 opened almost 2 years ago
by
Shuggu
Script to download all files of 1B sample data locally
👍
3
2
#13 opened almost 2 years ago
by
ivanzhouyq
[bot] Conversion to Parquet
#11 opened almost 2 years ago
by
parquet-converter
Request to provide 1B/10B/100B/1T token subsample datasets separately
2
#4 opened almost 2 years ago
by
johnhew
Want up to 10,000 FREE A40 hours to train models with? #GPUPOOR contest
#2 opened almost 2 years ago
by
rmonsurate
Open source community will forever be indebted to Together AI.
🤝
❤️
15
#1 opened almost 2 years ago
by
KrishnaKaasyap