|
|
e3e4b7abe6
|
Fix: Added code to allow for other data sources to be added
|
2024-06-08 09:21:19 -04:00 |
|
|
|
1fe54ed1ff
|
Fix: Adjusting Phoebe's code to prevent 'parroting'
|
2024-05-25 08:30:55 -04:00 |
|
|
|
509670c989
|
feat: Managed to achieve a loss of 0.285
|
2024-05-23 22:39:46 -04:00 |
|
|
|
47c8cce3dd
|
Fix: Working on improving the model code to get a better learning rate than 2.5
|
2024-05-17 23:33:02 -04:00 |
|
|
|
763514e815
|
Feat: Added a clean_data to process the data better
Feat: Added the new cleaned datasets
|
2024-05-17 14:15:44 -04:00 |
|
|
|
fb8db8a870
|
Fix: Working on the generate reply for discord.
Feat: Added a launch.json to allow quicker launches of the bot
docs: phoebe_model.pt will change every time we train.
|
2024-05-15 22:36:38 -04:00 |
|
|
|
75f1116b3b
|
Fix: Moved the Files around due to imports not working right
Feat: Phoebe replies but it's gibbish
This is a version break because of the file structure change.
|
2024-05-15 20:13:35 -04:00 |
|
|
|
12071fbf61
|
Feat!: Added train_gpt_model.py
This breaks any past code as it splits the code into two files.
doc: added phoebe_model.pt (trained model for phoebe)
|
2024-05-15 15:07:02 -04:00 |
|
|
|
54c4cf59b0
|
chore: added openwebtext and data_extract.py to the .gitignore
docs: added dataset
|
2024-05-15 12:46:19 -04:00 |
|
|
|
adca64bfc8
|
feat: Added GPT Model Code
Fix: Changed .pre-commit-confit.yaml to stop conflicts
docs: README.md changed due to the pre-commits
|
2024-05-14 21:15:36 -04:00 |
|
|
|
9db0796905
|
Whoops, Spelling Error
|
2024-05-07 08:21:25 -04:00 |
|
|
|
f33d1ddc62
|
First commit of everything
|
2024-05-06 20:58:44 -04:00 |
|
