Search results

Add check for loading old models

testsadding testsadding tests

Status: Open.

#211 In understanding-search/maze-transformer;

· mivanit opened on Mar 23, 2024

clean up notebooks

code-qualitycode quality improvementcode quality improvement

documentationImprovements or additions to documentationImprovements or additions to documentation

Status: Open.

#210 In understanding-search/maze-transformer;

· mivanit opened on Mar 5, 2024

stop integration tests from logging to wandb

bugSomething isn't workingSomething isn't working

code-qualitycode quality improvementcode quality improvement

wandbanything related to weights & biases integrationanything related to weights & biases integration

Status: Open.

#202 In understanding-search/maze-transformer;

· mivanit opened on Jan 9, 2024

streamline evals

code-qualitycode quality improvementcode quality improvement

evalsModel evaluationsModel evaluations

Status: Open.

#200 In understanding-search/maze-transformer;

· mivanit opened on Nov 16, 2023

·  

Fixing evals & training loop

investigate padding issues

bugSomething isn't workingSomething isn't working

questionFurther information is requestedFurther information is requested

researchResearch and ExperimentationResearch and Experimentation

Status: Open.

#196 In understanding-search/maze-transformer;

· mivanit opened on Oct 4, 2023

allow passing dataset to <code>train_model</code>

enhancementNew feature or requestNew feature or request

good first issueGood for newcomersGood for newcomers

Status: Open.

#194 In understanding-search/maze-transformer;

· mivanit opened on Oct 1, 2023

Calculate appropriate max sequence length based on the dataset

code-qualitycode quality improvementcode quality improvement

Status: Open.

#175 In understanding-search/maze-transformer;

· valedan opened on Apr 13, 2023

Allow dataset generation to take a seed

code-qualitycode quality improvementcode quality improvement

Status: Open.

#174 In understanding-search/maze-transformer;

· valedan opened on Apr 12, 2023

Automate direct logit attribution process

researchResearch and ExperimentationResearch and Experimentation

Status: Open.

#172 In understanding-search/maze-transformer;

· valedan opened on Apr 12, 2023

Show baseline performance in W&amp;B

evalsModel evaluationsModel evaluations

wandbanything related to weights & biases integrationanything related to weights & biases integration

Status: Open.

#171 In understanding-search/maze-transformer;

· valedan opened on Apr 12, 2023

·  

Fixing evals & training loop

Add eval for correct_forking_fraction

evalsModel evaluationsModel evaluations

Status: Open.

#167 In understanding-search/maze-transformer;

· valedan opened on Apr 12, 2023

·  

Fixing evals & training loop

Add logit-based evals

enhancementNew feature or requestNew feature or request

evalsModel evaluationsModel evaluations

Status: Open.

#165 In understanding-search/maze-transformer;

· valedan opened on Apr 12, 2023

·  

Fixing evals & training loop