Skip to content

Pull requests: EleutherAI/gpt-neox

Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Reviews
Assignee
Filter by who’s assigned
Sort

Pull requests list

Add Mamba Architecture
#1157 by haileyschoelkopf was merged Mar 10, 2024 Loading…
Add soft prompt tuning
#398 by sdtblck was merged Sep 8, 2021 Loading…
LR scheduler fix no longer breaks inference
#1060 by dashstander was merged Oct 17, 2023 Loading…
Lion Optimizer
#1062 by andylolu2 was merged Oct 20, 2023 Loading…
PR for Deepspeed Integration
#9 by trisongz was merged Dec 24, 2020 Loading…
get rid of test file
#10 by sdtblck was merged Dec 26, 2020 Loading…
make mask value smaller by factor of 2
#25 by lucidrains was merged Jan 4, 2021 Loading…
test
#1 by lucidrains was merged Dec 22, 2020 Loading…
Update base_model.json
#93 by srulikbd was merged Jan 26, 2021 Loading…
Implement distributed training using Kubernetes
#77 by leogao2 was merged Jan 23, 2021 Loading…
2
6
Batch size needs to be specified
#87 by joshlk was merged Jan 23, 2021 Loading…
Add checkpoint saving / loading
#90 by sdtblck was merged Jan 28, 2021 Loading…
Remove layer caching
#109 by joshlk was merged Feb 1, 2021 Loading…
Create runmany_k8s.sh
#106 by leogao2 was merged Jan 30, 2021 Loading…
Better deployment
#107 by joshlk was merged Feb 1, 2021 Loading…
Monitoring using wandb
#108 by joshlk was merged Feb 1, 2021 Loading…
Fix train pipeline
#89 by sdtblck was merged Jan 25, 2021 Loading…
2
1
Miscellaneous docker QoL improvements
#91 by leogao2 was merged Jan 26, 2021 Loading…
Minor fixes
#85 by sdtblck was merged Jan 23, 2021 Loading…
Update deploy_k8s.sh
#102 by leogao2 was merged Jan 27, 2021 Loading…
Update deploy_k8s.sh
#101 by leogao2 was merged Jan 27, 2021 Loading…
fix everything that i broke
#61 by sdtblck was merged Jan 13, 2021 Loading…
update tensorflow to 2.4.0
#47 by sdtblck was merged Jan 5, 2021 Loading…
ProTip! Type g i on any issue or pull request to go back to the issue listing page.