Datasets:

allenai
/

dolma

Tasks:

Text Generation

Languages:

English

Size:

n>1T

ArXiv:

Tags:

language-modeling

casual-lm

llm

License:

Dataset card Files Files and versions

xet

Community

New discussion

Resources

View closed (17)

Questions on Temporal Deduplication and Proposed Strategy in Dataset Construction

#52 opened 7 days ago by

AshleyLL

Inquiry on the Types of Documents Included in the Dataset

#51 opened 7 days ago by

AshleyLL

Availability of test/valid set

👍 2

#50 opened 4 months ago by

Parallaxixs

mismatch between the JSON dataset's schema

#48 opened 7 months ago by

losa-k

Gzip size of v1.7 is 4.1TB?

#47 opened 8 months ago by

rdiehlmartinez

How long time it takes to download or load the whole dataset?

#45 opened 9 months ago by

yliu2702

wget snippet wrong, files being overwritten/skipped

#44 opened 9 months ago by

effendietal

Llama v.s. OLMo token counts

#43 opened 10 months ago by

WillHeld

How many documents are actually in Dolma v1.7?

#42 opened 12 months ago by

yury-zyphra

Workaround for duplicated ID's issue

#41 opened 12 months ago by

yury-zyphra

Recommended wget snippet does not create folders

#40 opened 12 months ago by

yury-zyphra

Update dolma.py

👍 1

#39 opened 12 months ago by

kshinoda

Update README.md

👍 1

#38 opened 12 months ago by

kshinoda

Common Crawl Dataset Partitioning method?

#37 opened 12 months ago by

AlexFanWei

What is the total # tokens after sampling proportion? 1.7T or 1.65T

#36 opened about 1 year ago by

ivanzhouyq

missing shards in url.txt

#35 opened about 1 year ago by

Anony

v 1.7 Algebraic Stack is actually Proofpile's Arxiv?

#34 opened about 1 year ago by

Anony

Dolma1.7 books and c4 error

#33 opened about 1 year ago by

Chasell

Duplicates in Dolma 1.7

#32 opened about 1 year ago by

vedaad

Storage Footprint of Dolma 1.7 (on disk)

#31 opened about 1 year ago by

skaramcheti

Get values of `source` in dolma

👍 1

#30 opened about 1 year ago by

maxmatical

data_v_1_7 number of documents of dolma_cc

#29 opened about 1 year ago by

NICOLASORZ

Streaming access to the dataset raises an error

#26 opened about 1 year ago by

LorMolf

Dataset Viewer issue: DatasetWithScriptNotSupportedError

#25 opened about 1 year ago by

Balmerac

Please provide a list of file hashes in order to check integrity of downloads

#24 opened about 1 year ago by

markusheimerl

Question: How can I search the Dolma dataset without downloading it?

#21 opened over 1 year ago by

alishasrivas

Downloading the v1_6_sample data

#20 opened over 1 year ago by

ibenshaul

Duplicate Key when Loading Dataset

👍 1

#18 opened over 1 year ago by

Hhhhhao97

Data versions

#15 opened over 1 year ago by

Hudak

Metadata link missing

#12 opened over 1 year ago by

GokulNC

🚩 Report

#7 opened almost 2 years ago by

HiJohnHowsAsh

fix your licensing

#6 opened almost 2 years ago by

winddude

🚩 Report: Legal issue(s)

👍 5

#5 opened almost 2 years ago by

BarraHome

🚩 Report: Legal issue(s)

🤯 2

#3 opened almost 2 years ago by

endomorphosis

Compatibility of ImpACT MR license with wikipedia, project gutenberg etc

👍 6

#2 opened almost 2 years ago by

hac541309