issues
search
allenai
/
OLMo
Modeling, training, eval, and inference code for OLMo
https://allenai.org/olmo
Apache License 2.0
4.24k
stars
400
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Swap in Flan with correct separators.
#674
dwadden
opened
8 hours ago
0
mlp_ratio not adjusted in config if mlp_hidden_size is set
#673
Muennighoff
opened
1 day ago
0
Does global_train_batch_size support gradient accumulation?
#672
jinzhuoran
opened
1 day ago
0
Unshard S3 checkpoints and convert to HF.
#671
dwadden
opened
3 days ago
1
Fix restarts in later epochs
#670
epwalsh
closed
4 days ago
0
Log epoch # to console
#669
epwalsh
closed
4 days ago
0
update mamba train
#668
jacob-morrison
closed
5 days ago
0
change to mamba2
#667
jacob-morrison
closed
5 days ago
0
updating mamba branch to match main
#666
jacob-morrison
closed
5 days ago
0
sharded ckpt is saved only for fsdp
#665
ananyahjha93
closed
5 days ago
0
DDP training tries to save sharded checkpoint on the last step
#664
ananyahjha93
closed
5 days ago
0
Improving memmap type parser
#663
soldni
closed
5 days ago
0
updated config for olmo tiny suite of models
#662
ananyahjha93
closed
6 days ago
0
Add support for document masking during training
#661
epwalsh
closed
3 days ago
0
Add flag not to remove tmp files when doing hf conversion.
#660
dwadden
closed
4 days ago
0
Add option not to remove tmp directory when doing `convert_olmo_to_hf_new.py`
#659
dwadden
closed
6 days ago
1
Is there explicitly instruction-following data in the version of Dolma used to train v1?
#658
john-hewitt
opened
1 week ago
1
Update LUMI containers
#657
2015aroras
closed
1 week ago
0
Added = to CLI options on README.md
#656
jeqcho
closed
1 week ago
0
Can long text be splitted into short texts?
#655
CoinCheung
opened
1 week ago
0
Cannot convert internal OLMo checkpoint to HF
#654
viking-sudo-rm
opened
1 week ago
0
Bump version to v0.4.0
#653
2015aroras
closed
1 week ago
3
Update to torch 2.3
#652
2015aroras
closed
1 week ago
0
Update LUMI Dockerfile
#651
2015aroras
closed
1 week ago
0
start_index not getting reset in data loader when moving to new epoch
#650
leon-g-xu
opened
1 week ago
1
Model Ladder
#649
dirkgr
closed
4 days ago
0
Default to FSDP strategy
#648
2015aroras
closed
1 week ago
1
Use olmo_data tokenizer in Tokenizer.from_train_config
#647
2015aroras
closed
1 week ago
0
Load HF datasets from `olmo_data`
#646
2015aroras
closed
1 week ago
0
Move tokenizers to new `olmo_data` package.
#645
2015aroras
closed
2 weeks ago
1
Issue with tokenizer wrapper
#644
davidbrandfonbrener
opened
2 weeks ago
0
Fix off-by-one error in eval
#643
davidbrandfonbrener
closed
2 weeks ago
0
What did OLMo 1B converge to?
#642
sidereior
opened
2 weeks ago
0
Resuming training on unsharded checkpoint
#641
lecifire
opened
2 weeks ago
5
Multi node training
#640
shahizat
opened
2 weeks ago
0
MoE
#639
Muennighoff
opened
3 weeks ago
0
Scaling laws pipeline
#638
AkshitaB
opened
3 weeks ago
0
muP implementation
#637
AkshitaB
opened
3 weeks ago
0
Add option to skip optim steps for 0 grad params
#636
epwalsh
closed
1 week ago
0
Unit tests
#635
AkshitaB
opened
3 weeks ago
0
Fix Z-loss calculation
#634
epwalsh
closed
3 weeks ago
4
Tokenizer with relative path import fails when using olmo as pip library
#633
viking-sudo-rm
closed
2 weeks ago
0
How the 1B and 7B model are initialized?
#632
sanyalsunny111
opened
4 weeks ago
0
Make hf_olmo support AutoModelForCausalLM
#631
2015aroras
closed
3 weeks ago
0
Amberish 7B hero run
#629
epwalsh
opened
1 month ago
0
Olmo tiny scripts
#628
ananyahjha93
closed
3 weeks ago
0
Key 'https://olmo_checkpoints' not in 'TrainConfig'
#627
jeqcho
closed
1 week ago
1
Inspect training data improvements
#626
2015aroras
closed
1 month ago
0
What is the true MLP ratio for OLMo 7B?
#625
jeqcho
closed
1 month ago
2
Make olmo-core checkpointer more robust on weka
#624
epwalsh
closed
1 month ago
0
Next