Skip to content

Issues: minimaxir/gpt-2-simple

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

Clear current search query, filters, and sorts
0 Open 101 Closed
0 Open 101 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

Text is incoherent when generating with batch_size > 1
#1 by minimaxir was closed Apr 17, 2019
1
Anyone understand the code of the small model? I'm looking to hire someone!
#4 by bladedsupernova was closed Apr 20, 2019
1
The first word of the prefix is trimmed to 1 character bug Something isn't working
#5 by woctezuma was closed Jun 15, 2019
6
Fails to load dataset on Windows due to text encoding bug Something isn't working
#9 by bob80333 was closed Apr 21, 2019
7
Error when calling copy_checkpoint_to_gdrive()
#11 by woctezuma was closed Apr 21, 2019
1
ValueError: Variable model/wpe already exists, disallowed
#12 by woctezuma was closed Apr 21, 2019
1
Too many iterations when fine-tuning a model a second time
#13 by woctezuma was closed Apr 21, 2019
3
Finetuning: fresh vs. latest
#14 by woctezuma was closed Apr 21, 2019
4
Other examples using gpt-2-simple
#16 by woctezuma was closed Apr 21, 2019
1
Argument for copy_checkpoint_to_gdrive() and copy_checkpoint_from_gdrive()
#17 by woctezuma was closed May 18, 2019
Ability encode files to tokens separately from fine-tuning?
#19 by bob80333 was closed Jun 19, 2019
5
Add parameter to keep only one checkpoint while restore_from='latest' enhancement New feature or request
#20 by minimaxir was closed May 20, 2019 v0.5
1
One-off counter when checkpoint saves bug Something isn't working
#21 by minimaxir was closed May 5, 2019
Update for 345M enhancement New feature or request
#23 by david-sg was closed May 5, 2019
4
Number of tokens on file imported, 4x actual?
#25 by david-sg was closed May 5, 2019
3
345M Directory Issue
#26 by WAUthethird was closed May 6, 2019
2
Loss
#27 by david-sg was closed May 6, 2019
Error when I try to generate text from finetuned model
#28 by jesse-lane-ai was closed May 7, 2019
1
Enhancement - Can you make the default readme generated text from terminator scripts?
#29 by johndpope was closed May 7, 2019
3
0 tokens when attempting to finetune using .txt
#31 by lukegalea was closed Jul 9, 2019
7
How do I evaluate perplexity of the lm
#32 by ghost was closed May 8, 2019
1
Importance of prepending <|startoftext|> and suffixing <|endoftext|>
#33 by woctezuma was closed May 8, 2019
3
Saving checkpoints to Google Drive enhancement New feature or request
#34 by woctezuma was closed May 18, 2019
4
Default temperature
#43 by woctezuma was closed May 11, 2019
2
Not an issue - fyi / controllable neural plot generation
#44 by johndpope was closed Oct 2, 2019
ProTip! Adding no:label will show everything without a label.