Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / karpathy/ng-video-lecture issues and pull requests
#48 - generated gibberish within the block
Issue -
State: open - Opened by yangyangyyy123 about 1 month ago
#47 - `FeedFoward` to `FeedForward` [typo] gpt.py
Pull Request -
State: open - Opened by Vasudeva-bit about 2 months ago
#46 - Shouldn't we be dividing when normalizing QK^T, not multiplying?
Issue -
State: open - Opened by tylerkastner 2 months ago
#45 - refactor: linting
Pull Request -
State: closed - Opened by shivakumarmahesh 3 months ago
#44 - No MIT license file in the repository
Issue -
State: open - Opened by matthewcarbone 4 months ago
#43 - The mathematical trick in self-attention, why it returns false for torch.allclose(xbow, xbow2)?
Issue -
State: open - Opened by Ryan-ZL-Lin 9 months ago
- 2 comments
#42 - Strange model behavior when taking the softmax in the wrong dimension
Issue -
State: open - Opened by Cloud299 9 months ago
- 1 comment
#41 - disabled gradient calculation in generate function (bigram.py)
Pull Request -
State: open - Opened by arun477 10 months ago
- 1 comment
#40 - may you share code to run only inference
Issue -
State: open - Opened by Sandy4321 10 months ago
- 2 comments
#39 - supplementary video lecture: may you share link to this video pls
Issue -
State: open - Opened by Sandy4321 10 months ago
- 2 comments
#38 - can it be run on ubuntu PC with nvidia 3060 GPU 8 GB
Issue -
State: open - Opened by Sandy4321 10 months ago
- 2 comments
#37 - can be windows OS with only CPU used ?
Issue -
State: open - Opened by Sandy4321 10 months ago
- 4 comments
#36 - Using the variable "model" after declaring variable "m"
Issue -
State: closed - Opened by klivin 11 months ago
- 1 comment
#35 - mac studio can't generate token
Issue -
State: open - Opened by arthasyou about 1 year ago
- 2 comments
#34 - How is torch broadcasting (T, T) @ (B, T, C) ?!
Issue -
State: open - Opened by whydna about 1 year ago
- 4 comments
#33 - Dev
Pull Request -
State: closed - Opened by jhancock1975 about 1 year ago
- 2 comments
#32 - KeyError
Issue -
State: closed - Opened by yihaoye about 1 year ago
#31 - gpt.py how to save the model after training and how to use it so that it returns the text to me relevant to ChatGPT?
Issue -
State: open - Opened by MrKsiJ about 1 year ago
- 5 comments
#30 - wei value not 100% per row after dropout
Issue -
State: open - Opened by guyko81 about 1 year ago
- 1 comment
#29 - how to save, Load and Finetune the model
Issue -
State: open - Opened by Lokeshwaran-M about 1 year ago
- 1 comment
#28 - Discrepancy with dimensions
Issue -
State: open - Opened by BasedLukas over 1 year ago
#27 - edit shape comments in `generate` method
Pull Request -
State: closed - Opened by cthiriet over 1 year ago
#26 - M1/M2 performance fix: use Apple MPS (metal performance shaders) if available
Pull Request -
State: open - Opened by sghael over 1 year ago
- 3 comments
#25 - Change the Title Please
Issue -
State: open - Opened by kentonbmax over 1 year ago
#24 - Position embedding seems wrong
Issue -
State: closed - Opened by zurtal over 1 year ago
#23 - Might want to modify README to remove the "NOTE"
Issue -
State: open - Opened by zjmiller over 1 year ago
#22 - Minor correction in 'Add & Norm' logic in Block Class in gpt.py
Pull Request -
State: open - Opened by AbhishekAshokDubey over 1 year ago
- 1 comment
#21 - About gpt.py line 134-135
Issue -
State: open - Opened by hufuzhipeng over 1 year ago
- 1 comment
#20 - "index out of range" error when using a different embedding dimension than vocab_size
Issue -
State: open - Opened by zhoupingjay over 1 year ago
- 1 comment
#19 - Merge #1
Pull Request -
State: closed - Opened by yrraadi-io over 1 year ago
#18 - bug?: m vs model
Issue -
State: open - Opened by freestylerick over 1 year ago
- 6 comments
#17 - updated dictionary names
Pull Request -
State: closed - Opened by caaker over 1 year ago
- 1 comment
#16 - Call `model.eval()` before generating?
Issue -
State: open - Opened by gustavdelius over 1 year ago
- 1 comment
#15 - No license file
Issue -
State: open - Opened by Maniues over 1 year ago
- 1 comment
#14 - time series data like BTC price
Issue -
State: closed - Opened by tesla-cat over 1 year ago
- 1 comment
#13 - Adding architecture diagram for nanogpt
Pull Request -
State: open - Opened by patchy631 over 1 year ago
#12 - adding option to use the bpe tokenizer tiktoken as mentioned in the lecture
Pull Request -
State: closed - Opened by jhlimm8 over 1 year ago
- 2 comments
#11 - Teeny-tiny performance improvement
Pull Request -
State: open - Opened by Andrei-Aksionov almost 2 years ago
#10 - Loss calculation should not permanently change shapes of logits and targets
Pull Request -
State: open - Opened by Andrei-Aksionov almost 2 years ago
#9 - Clarify shape descriptions inside forward method
Pull Request -
State: closed - Opened by Andrei-Aksionov almost 2 years ago
- 2 comments
#8 - Add Keras Counterparts
Pull Request -
State: open - Opened by j-planet almost 2 years ago
#7 - removes unnecessary duplicate variable
Pull Request -
State: open - Opened by hoosha almost 2 years ago
- 1 comment
#6 - We should scale attention by head size
Pull Request -
State: closed - Opened by vineetm almost 2 years ago
- 1 comment
#5 - no longer bigram model?
Issue -
State: open - Opened by eniompw almost 2 years ago
- 1 comment
#3 - Automating data download if needed
Pull Request -
State: open - Opened by nicholas-dinicola almost 2 years ago
#2 - UML diagram helping beginners understand gpt.py
Issue -
State: open - Opened by QasimWani almost 2 years ago
- 2 comments
#1 - fix linting errors
Pull Request -
State: open - Opened by mendi80 almost 2 years ago