Play project
Investigating Training Dynamics via Token Loss Trajectories's itch.io pageResults
Criteria | Rank | Score* | Raw Score |
Novelty | #1 | 4.667 | 4.667 |
Generality | #1 | 3.667 | 3.667 |
Reproducibility | #1 | 4.333 | 4.333 |
Safety | #2 | 3.333 | 3.333 |
Judge's choice | #2 | n/a | n/a |
Benchmark | #3 | 3.000 | 3.000 |
Ranked from 3 ratings. Score is adjusted from raw score by the median number of ratings per game in the jam.
Judge feedback
Judge feedback is anonymous.
- This is really good work in testing for locations to look at using interpretability tools! Further developing this idea into something that can be used as a testing suite would be quite interesting and the ideas from the project seem useful at scale.
Where did you participate?
Online
What are the full names of the participants?
Alex Foote
Leave a comment
Log in with itch.io to leave a comment.
Comments
No one has posted a comment yet