Skip to content

Issues: HomebrewNLP/Olmax

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

26 Open 21 Closed
26 Open 21 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Multiple forward per backward core Improves core model while keeping core idea intact engineering Software-engineering problems that don't require ML-Expertise research Creative project that might fail but could give high returns
#81 opened Sep 11, 2022 by ClashLuke
Staged batchsize training core Improves core model while keeping core idea intact engineering Software-engineering problems that don't require ML-Expertise research Creative project that might fail but could give high returns
#80 opened Sep 11, 2022 by ClashLuke
Compact Loss core Improves core model while keeping core idea intact research Creative project that might fail but could give high returns
#79 opened Sep 11, 2022 by ClashLuke
Causality Test core Improves core model while keeping core idea intact engineering Software-engineering problems that don't require ML-Expertise
#78 opened Sep 11, 2022 by ClashLuke
1
Square LR-Schedule core Improves core model while keeping core idea intact engineering Software-engineering problems that don't require ML-Expertise
#70 opened Aug 13, 2022 by ClashLuke
Alternative Losses core Improves core model while keeping core idea intact ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#54 opened May 24, 2022 by ClashLuke
2
Long-Range-Arena Evaluation downstream Changes code wrapping the core model engineering Software-engineering problems that don't require ML-Expertise ML Requires machine-learning knowledge (can be built up on the fly)
#49 opened May 18, 2022 by ClashLuke
Gradient Noise core Improves core model while keeping core idea intact ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#46 opened May 17, 2022 by ClashLuke
Retrieval Augmented Causal Generation core Improves core model while keeping core idea intact ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#45 opened May 17, 2022 by ClashLuke
Encoder-Decoder Architecture core Improves core model while keeping core idea intact ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#44 opened May 17, 2022 by ClashLuke
Automated Eval-Demo Update engineering Software-engineering problems that don't require ML-Expertise mlops
#32 opened May 9, 2022 by ClashLuke
@ma7dev
Automated Long-Running Experiments engineering Software-engineering problems that don't require ML-Expertise mlops
#31 opened May 9, 2022 by ClashLuke
@ma7dev
Automated Integration Tests engineering Software-engineering problems that don't require ML-Expertise mlops
#30 opened May 9, 2022 by ClashLuke
@ma7dev
Long-Context Experiments engineering Software-engineering problems that don't require ML-Expertise ML Requires machine-learning knowledge (can be built up on the fly)
#29 opened May 9, 2022 by ClashLuke First Release
"Resume" option for tokenizers downstream Changes code wrapping the core model engineering Software-engineering problems that don't require ML-Expertise
#23 opened Apr 30, 2022 by ClashLuke
Release pretrained weights
#22 opened Apr 30, 2022 by ClashLuke First Release
Language-Model Evaluation downstream Changes code wrapping the core model engineering Software-engineering problems that don't require ML-Expertise ML Requires machine-learning knowledge (can be built up on the fly)
#21 opened Apr 30, 2022 by ClashLuke First Release
Frontend engineering Software-engineering problems that don't require ML-Expertise
#20 opened Apr 30, 2022 by ClashLuke First Release
@ma7dev
Stabilize MoE core Improves core model while keeping core idea intact engineering Software-engineering problems that don't require ML-Expertise ML Requires machine-learning knowledge (can be built up on the fly)
#16 opened Apr 30, 2022 by ClashLuke
Non-Autoregressive Generation downstream Changes code wrapping the core model ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#12 opened Apr 30, 2022 by ClashLuke
2
Image Classification downstream Changes code wrapping the core model ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#11 opened Apr 30, 2022 by ClashLuke
Tokenizing Phonetics downstream Changes code wrapping the core model ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#10 opened Apr 30, 2022 by ClashLuke
7
Audio Modelling ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#9 opened Apr 30, 2022 by ClashLuke
Explicit Memory core Improves core model while keeping core idea intact ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#8 opened Apr 30, 2022 by ClashLuke
MoE + Weight Sharing core Improves core model while keeping core idea intact ML Requires machine-learning knowledge (can be built up on the fly) research Creative project that might fail but could give high returns
#6 opened Apr 30, 2022 by ClashLuke
ProTip! Follow long discussions with comments:>50.