Maatra

Went through the warning list from pytest and tryed to fix as many warning as i could that should not be raised within testing. This allows relevant warnings to not...

moved image diff logging to their respective workspace

1

#809 was flawed and didn't change the programming logic. I tried to think of any way for build_workspace_and_update_role() to return whether a change in the built image is detected. There...

Docker build

2

Added a docker API client, and changed the build function to its low-level version. This makes it return an event stream that we can then log to screen for real-time...

Docker build verbosity

3

## Description Changing the docker image build to its low level implementation so it can be more verbose. ## Motivation/Background Building the docker image can take quite some time, and...

Hyperparameter tracking

12

Sensitive parameters are not really DRY since they are repeated in verify_command_line_training_args, but for now only two repeats is not too bad. The setting of hyperparameters through config is handled...

Tokenization for the large model

2

1) The config for large model specifies a vocab size of 51200, is there a separate tokenizer file for it? Weirdly vocab falls back down to 32 for xlarge which...