Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / karpathy/minGPT issues and pull requests
#143 - Reallly small correction
Pull Request -
State: open - Opened by manuelsh about 1 month ago
#140 - <|endoftext|> token isn't encoded correctly
Issue -
State: open - Opened by ttumiel 5 months ago
- 2 comments
#138 - DEMO: minGPT on tinygrad
Pull Request -
State: open - Opened by ziliangpeng 7 months ago
#100 - fix: add missing dependency in `setup.py`
Pull Request -
State: open - Opened by ben-schulz almost 2 years ago
#99 - tests do not run in project as built
Issue -
State: open - Opened by ben-schulz almost 2 years ago
- 1 comment
#98 - Facilitating setup with popular tools
Issue -
State: open - Opened by Utopiah almost 2 years ago
#97 - Fix typo in bpe.py
Pull Request -
State: open - Opened by eltociear almost 2 years ago
#97 - Fix typo in bpe.py
Pull Request -
State: open - Opened by eltociear almost 2 years ago
#96 - Update README.md
Pull Request -
State: closed - Opened by chinhaihour almost 2 years ago
#96 - Update README.md
Pull Request -
State: closed - Opened by chinhaihour almost 2 years ago
#95 - Caching for generation
Issue -
State: open - Opened by murbard almost 2 years ago
- 1 comment
#95 - Caching for generation
Issue -
State: open - Opened by murbard almost 2 years ago
- 1 comment
#94 - Renaming transformer.h into transformer.l
Issue -
State: open - Opened by marxav almost 2 years ago
#93 - Wrong definition of Query, Key, Value matrices? They shouldn't have bias=True
Issue -
State: open - Opened by LeoPerelli almost 2 years ago
- 3 comments
#92 - Remove extraneous semicolons, remind users to get transformers repo if needed
Pull Request -
State: open - Opened by danielgross about 2 years ago
#92 - Remove extraneous semicolons, remind users to get transformers repo if needed
Pull Request -
State: open - Opened by danielgross about 2 years ago
#91 - `named_parameters` does not have to be recursive
Pull Request -
State: open - Opened by Equim-chan about 2 years ago
#91 - `named_parameters` does not have to be recursive
Pull Request -
State: open - Opened by Equim-chan about 2 years ago
#90 - cannot import name 'sample' from 'mingpt.utils'
Issue -
State: open - Opened by chris-the-wiz about 2 years ago
- 1 comment
#89 - Update readme.md
Pull Request -
State: open - Opened by macleginn about 2 years ago
#89 - Update readme.md
Pull Request -
State: open - Opened by macleginn about 2 years ago
#88 - Update README.md
Pull Request -
State: open - Opened by macleginn about 2 years ago
#87 - Add DataParallel and make Block support DataParallel
Pull Request -
State: open - Opened by gngdb about 2 years ago
#86 - Add dtype support
Pull Request -
State: open - Opened by younesbelkada over 2 years ago
#86 - Add dtype support
Pull Request -
State: open - Opened by younesbelkada over 2 years ago
#85 - Move token + pos embedding computation to a separate method
Pull Request -
State: closed - Opened by ericjang over 2 years ago
- 12 comments
#85 - Move token + pos embedding computation to a separate method
Pull Request -
State: closed - Opened by ericjang over 2 years ago
- 12 comments
#84 - Add setup.py to allow mingpt to be used as a third-party library
Pull Request -
State: closed - Opened by ericjang over 2 years ago
- 1 comment
#83 - Use XOR operator `^` for checking assertion `type_given XOR params_given`
Pull Request -
State: closed - Opened by mishig25 over 2 years ago
- 1 comment
#82 - Fix README.md typo
Pull Request -
State: closed - Opened by neverix over 2 years ago
- 1 comment
#81 - Add optimizer to Trainer's self for callbacks.
Pull Request -
State: closed - Opened by luigidisotto over 2 years ago
- 1 comment
#80 - Refactor for modern 2022 python style and usage
Pull Request -
State: open - Opened by mattsta over 2 years ago
- 3 comments
#80 - Refactor for modern 2022 python style and usage
Pull Request -
State: open - Opened by mattsta over 2 years ago
- 3 comments
#79 - Is it more reasonable to only use causal attention in the first block of GPT
Issue -
State: open - Opened by charlesxu90 over 2 years ago
#79 - Is it more reasonable to only use causal attention in the first block of GPT
Issue -
State: open - Opened by charlesxu90 over 2 years ago
#78 - Typos
Pull Request -
State: closed - Opened by nat over 2 years ago
- 1 comment
#78 - Typos
Pull Request -
State: closed - Opened by nat over 2 years ago
- 1 comment
#77 - Meaning of "-1 because very last digit doesn't plug back"
Issue -
State: open - Opened by vwxyzjn over 2 years ago
#77 - Meaning of "-1 because very last digit doesn't plug back"
Issue -
State: open - Opened by vwxyzjn over 2 years ago
#76 - Refactor repo into script-based projects
Pull Request -
State: closed - Opened by karpathy over 2 years ago
#76 - Refactor repo into script-based projects
Pull Request -
State: closed - Opened by karpathy over 2 years ago
#75 - Perfect training and evaluation loss, but terrible test-time performance
Issue -
State: closed - Opened by micahcarroll over 2 years ago
- 1 comment
#75 - Perfect training and evaluation loss, but terrible test-time performance
Issue -
State: closed - Opened by micahcarroll over 2 years ago
- 1 comment
#74 - Add unit tests
Pull Request -
State: closed - Opened by mishig25 over 2 years ago
- 1 comment
#74 - Add unit tests
Pull Request -
State: closed - Opened by mishig25 over 2 years ago
- 1 comment
#73 - Add tests
Pull Request -
State: closed - Opened by mishig25 over 2 years ago
#73 - Add tests
Pull Request -
State: closed - Opened by mishig25 over 2 years ago
#72 - #71 use config n_head instead of hardcoded 4 heads
Pull Request -
State: open - Opened by SpeedCoder5 over 2 years ago
#72 - #71 use config n_head instead of hardcoded 4 heads
Pull Request -
State: open - Opened by SpeedCoder5 over 2 years ago
#71 - model self-attention hardcoded to 4 heads
Issue -
State: closed - Opened by SpeedCoder5 over 2 years ago
- 1 comment
#70 - How to handle unequal sequence length in a batch
Issue -
State: closed - Opened by luxuantao over 2 years ago
- 2 comments
#69 - Added the condition for test_dataset's presence.
Pull Request -
State: open - Opened by RohanAwhad over 2 years ago
#69 - Added the condition for test_dataset's presence.
Pull Request -
State: open - Opened by RohanAwhad over 2 years ago
#68 - Integration with HuggingFace
Issue -
State: open - Opened by marxav over 2 years ago
#68 - Integration with HuggingFace
Issue -
State: open - Opened by marxav over 2 years ago
#67 - play_math AdditionDataset.__get_item__ return value?
Issue -
State: open - Opened by SpeedCoder5 over 2 years ago
#67 - play_math AdditionDataset.__get_item__ return value?
Issue -
State: open - Opened by SpeedCoder5 over 2 years ago
#66 - Add distributed data parallel trainer
Pull Request -
State: open - Opened by aravindsrinivas over 2 years ago
#65 - How to determine `warmup_tokens` and `final_tokens`?
Issue -
State: open - Opened by fgolemo almost 3 years ago
#64 - modify to use pooling instead
Pull Request -
State: closed - Opened by annasajkh almost 3 years ago
#63 - Fix broken hugging face link & add link to huggingface / transformers
Pull Request -
State: closed - Opened by mishig25 almost 3 years ago
- 2 comments
#63 - Fix broken hugging face link & add link to huggingface / transformers
Pull Request -
State: closed - Opened by mishig25 almost 3 years ago
- 2 comments
#62 - initialize position embeddings
Pull Request -
State: closed - Opened by t-vi about 3 years ago
- 1 comment
#62 - initialize position embeddings
Pull Request -
State: closed - Opened by t-vi about 3 years ago
- 1 comment
#61 - Code completion notebook
Pull Request -
State: closed - Opened by tetelestia over 3 years ago
- 2 comments
#60 - Implemented play_word.ipynb example
Pull Request -
State: open - Opened by emukans over 3 years ago
- 1 comment
#59 - Error when I provide test dataset (custom minGPT)
Issue -
State: open - Opened by asigalov61 over 3 years ago
#59 - Error when I provide test dataset (custom minGPT)
Issue -
State: open - Opened by asigalov61 over 3 years ago
#58 - minor typo
Pull Request -
State: closed - Opened by project-delphi over 3 years ago
- 1 comment
#58 - minor typo
Pull Request -
State: closed - Opened by project-delphi over 3 years ago
- 1 comment
#57 - TPU/GPU training: KeyError 'pos_emb'
Issue -
State: open - Opened by tech509201941 over 3 years ago
#56 - How do I see test loss?
Issue -
State: open - Opened by aletote almost 4 years ago
- 1 comment
#55 - perform sqrt-d scaling on q instead of att matrix
Pull Request -
State: closed - Opened by aravindsrinivas almost 4 years ago
- 2 comments
#55 - perform sqrt-d scaling on q instead of att matrix
Pull Request -
State: closed - Opened by aravindsrinivas almost 4 years ago
- 2 comments
#54 - Question about memory usage for play_math
Issue -
State: open - Opened by pablogranolabar almost 4 years ago
#54 - Question about memory usage for play_math
Issue -
State: open - Opened by pablogranolabar almost 4 years ago
#53 - move instantiation of DataLoader
Pull Request -
State: closed - Opened by waynemystir almost 4 years ago
- 1 comment
#53 - move instantiation of DataLoader
Pull Request -
State: closed - Opened by waynemystir almost 4 years ago
- 1 comment
#52 - Layer norm should be after residual block
Pull Request -
State: closed - Opened by Scikud about 4 years ago
- 1 comment
#52 - Layer norm should be after residual block
Pull Request -
State: closed - Opened by Scikud about 4 years ago
- 1 comment
#51 - Use of amp.autocast does not improve performance
Issue -
State: open - Opened by aurotripathy about 4 years ago
- 1 comment
#51 - Use of amp.autocast does not improve performance
Issue -
State: open - Opened by aurotripathy about 4 years ago
- 1 comment
#50 - Add a multi-language example
Pull Request -
State: closed - Opened by marxav about 4 years ago
#50 - Add a multi-language example
Pull Request -
State: closed - Opened by marxav about 4 years ago
#49 - Add a "How to cite this work" section in README.md
Issue -
State: open - Opened by marxav about 4 years ago
- 2 comments
#49 - Add a "How to cite this work" section in README.md
Issue -
State: open - Opened by marxav about 4 years ago
- 2 comments
#48 - How to apply to time series?
Issue -
State: closed - Opened by thinkingparticle about 4 years ago
- 1 comment
#47 - Will this repo add the reformer or teach people how to implement reformer in Pytorch?
Issue -
State: closed - Opened by JonathanSum about 4 years ago
- 2 comments
#46 - If possible, please add TQDM (auto)
Pull Request -
State: closed - Opened by asigalov61 about 4 years ago
- 5 comments
#46 - If possible, please add TQDM (auto)
Pull Request -
State: closed - Opened by asigalov61 about 4 years ago
- 5 comments
#45 - Is there a Tensorflow-version of minGPT? Or will someone implement it?
Issue -
State: closed - Opened by guotong1988 about 4 years ago
- 2 comments
#45 - Is there a Tensorflow-version of minGPT? Or will someone implement it?
Issue -
State: closed - Opened by guotong1988 about 4 years ago
- 2 comments
#44 - GPT vs BERT, under same computation and data resource, which one is better for downstream tasks like GLUE?
Issue -
State: open - Opened by guotong1988 about 4 years ago
- 1 comment
#44 - GPT vs BERT, under same computation and data resource, which one is better for downstream tasks like GLUE?
Issue -
State: open - Opened by guotong1988 about 4 years ago
- 1 comment
#43 - #39 Adding B, T, C description in comment.
Pull Request -
State: closed - Opened by JonathanSum about 4 years ago
- 1 comment
#43 - #39 Adding B, T, C description in comment.
Pull Request -
State: closed - Opened by JonathanSum about 4 years ago
- 1 comment
#42 - Sharing Pretrained Checkpoints
Issue -
State: open - Opened by barisbatuhan about 4 years ago
- 2 comments
#41 - typo 'terations'
Pull Request -
State: closed - Opened by JonathanSum about 4 years ago
- 1 comment
#41 - typo 'terations'
Pull Request -
State: closed - Opened by JonathanSum about 4 years ago
- 1 comment
#40 - Potential encoding issue in addition problem in play_math notebook?
Issue -
State: closed - Opened by ravi-annaswamy about 4 years ago
- 3 comments