Maurice Weber
mauriceweber
AI & ML interests
None yet
Organizations
mauriceweber's activity
RPV2 ccnet preprocessing
1
#29 opened about 1 month ago
by
bpwl0121
sample split details
3
#4 opened 11 months ago
by
sujantkumarkv
How can I download the sample-10B fastestly?
1
#28 opened 3 months ago
by
zgxiao
defunct book subset
4
#28 opened 11 months ago
by
polinaeterna
How much disk space would the whole HF dataset take?
1
#27 opened 6 months ago
by
protossw512
rpv2-subsamples
1
#26 opened 9 months ago
by
mauriceweber
The doc_id in duplicates is should contain?
3
#24 opened 9 months ago
by
newbietuan
Deduplication steps
23
#15 opened 11 months ago
by
ilyayudkovich
Here's a download script parallelized using Spark
1
#22 opened 10 months ago
by
srowen
what is the meaning of snapshots in redpajama-data-v2?
2
#21 opened 10 months ago
by
choidonghun
How to join documents and quality signals when downloading directly
3
#19 opened 10 months ago
by
tgshdyfuhuf
Missing duplicates parquet files
5
#18 opened 10 months ago
by
bebensee
Script to download all files of 1B sample data locally
2
#13 opened 11 months ago
by
ivanzhouyq
What is the total size, of the entirety of this dataset in TB?
1
#10 opened 11 months ago
by
Bayaz
What's the concept on partitions
2
#5 opened 11 months ago
by
SwatCat
quality_signals, minhash and duplicates missing
2
#3 opened 11 months ago
by
sheshanshag
Request to add retries into RedPajama-Data-V2.py script
1
#16 opened 10 months ago
by
yura38
How to obtain duplicates from minhash?
1
#8 opened 11 months ago
by
cq
Obtaining Filtered Samples
4
#12 opened 11 months ago
by
ssingh22
How big is the data size of en?
5
#6 opened 11 months ago
by
newbietuan
Request to provide 1B/10B/100B/1T token subsample datasets separately
2
#4 opened 11 months ago
by
johnhew
Missing file error
3
#9 opened 11 months ago
by
emrgnt-cmplxty
RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'
4
#29 opened about 1 year ago
by
shubhamagarwal92
The model doesn't seem to stop
15
#1 opened about 1 year ago
by
LaferriereJC
llama2 forward pass seemingly not working with padded inputs, unless one element in batch is not padded
3
#13 opened 12 months ago
by
joehakim
Using the Accelerate API to train models on multiple GPUs
8
#28 opened about 1 year ago
by
ajash
Input validation error: `max_new_tokens` must be <= 1. Given: 20
1
#12 opened about 1 year ago
by
reubenlee3
Keep getting error while loading tokenizer = AutoTokenizer.from_pretrained("togethercomputer/LLaMA-2-7B-32K")
5
#27 opened about 1 year ago
by
AIHero123
Are the unsafe files from C4 also in RedPajama?
2
#26 opened about 1 year ago
by
cwallenwein
Prompt format different in dataset compared to model card
3
#11 opened about 1 year ago
by
bhperry
Model gives itself instructions and keeps going and going and going?
5
#8 opened about 1 year ago
by
michael-newsrx-com
Great model. Plans for 13b version?
1
#9 opened about 1 year ago
by
nahuel89p
Loading model without fast-attn
1
#10 opened about 1 year ago
by
TZ20
Model on your API Playground
7
#3 opened about 1 year ago
by
1littlecoder
Can I continue pretraining this model for domain adaptation?
4
#6 opened about 1 year ago
by
sadahila
inconsistent data field in github jsonl files
3
#24 opened about 1 year ago
by
Rita
Unwanted repetitive response
3
#12 opened about 1 year ago
by
sdranju
protofile.proto: A file with this name is already in the pool
1
#19 opened about 1 year ago
by
surya-narayanan
ENDPOINT CONFIGURATION ON AWS SAGEMAKER
1
#21 opened about 1 year ago
by
NABARKA
Any plans for chat model?
1
#5 opened about 1 year ago
by
brekk
when will have a ggml version?
8
#3 opened about 1 year ago
by
CUIGuy
Skip split generation.
3
#23 opened about 1 year ago
by
luosuu
LocalAI Model Loading
3
#2 opened about 1 year ago
by
FIWisher
Error when loading book/book.jsonl using load_dataset
5
#22 opened about 1 year ago
by
icycold
Instead of flash_attn it should be flash_attn_2_cuda . This is causing a deployment issue in TGI/DJL
1
#14 opened about 1 year ago
by
monuminu
!pip install flash-attn --no-build-isolation
2
#15 opened about 1 year ago
by
NivYO
getting strange tokens after finetuning on Qlora
2
#11 opened about 1 year ago
by
monuminu
RoPE scaling and max_position_embeddings
2
#12 opened about 1 year ago
by
ag0
What is the VRAM requirement of this model?
5
#1 opened about 1 year ago
by
Said2k
GGML Version
8
#4 opened about 1 year ago
by
s3nh
Can try code as long text data.
1
#1 opened about 1 year ago
by
win10
Training diverges when used with Llama 2 70B and 4-bit QLoRA
3
#10 opened about 1 year ago
by
alyssavance
Specify RLHF data for the Instruct and Chat versions in model card
3
#9 opened about 1 year ago
by
markding
What's the prompt template?
11
#4 opened over 1 year ago
by
qiz
Is this model commercially usable?
2
#10 opened over 1 year ago
by
AayushShah