Skip to content

Issues: bigscience-workshop/training_dynamics

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

11 Open 0 Closed
11 Open 0 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

Design initial probes
#12 opened Apr 15, 2022 by rteehas
Set up infrastructure to isolate specific layers or groups of layers in the network for probing
#11 opened Apr 15, 2022 by rteehas
@oserikov
Extract the dataset or data generating process from (Yao et. al. 2021)
#10 opened Apr 15, 2022 by rteehas
@rteehas
Implement auxiliary prediction tasks (from: Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks)
#9 opened Jan 28, 2022 by oserikov
2 tasks
Implement saliency methods (from: The elephant in the interpretability room: Why use attention as explanation when we have saliency methods?)
#8 opened Jan 28, 2022 by oserikov
2 tasks
Implement the control tasks technique to ensure probes are selective (from Designing and Interpreting Probes with Control Tasks)
#7 opened Jan 22, 2022 by oserikov
2 tasks
Implement the SentEval methods (linear and MLP; from What you can cram into a single {\$}{\&}!{\#}* vector: Probing sentence embeddings for linguistic properties) good first issue Good for newcomers
#6 opened Jan 22, 2022 by oserikov
2 tasks
Implement differentiable weight masks for modularity detection (from Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks) good first issue Good for newcomers
#4 opened Aug 31, 2021 by rteehas
@hakunanatasha
1
Implement the parse depth structural probe (from A Structural Probe for Finding Syntax in Word Representations) good first issue Good for newcomers
#3 opened Aug 31, 2021 by rteehas
@wadimiusz
1
Test initial structural probes on BERT to replicate Manning and Hewitt results
#2 opened Aug 31, 2021 by rteehas
1
Implement the structural probe for syntax structure in Manning and Hewitt 2019 good first issue Good for newcomers
#1 opened Aug 31, 2021 by rteehas
@oserikov
4
ProTip! no:milestone will show everything without a milestone.