Skip to content

Pull requests: karpathy/llm.c

New pull request New
99 Open 476 Closed
99 Open 476 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Adamw thread coarsening kernel
#753 opened Sep 3, 2024 by saladpalad Loading…
implement rmsnorm in C
#750 opened Aug 28, 2024 by Jake-Song Loading…
Fix sizing typo in train_gpt2_fp32.cu
#748 opened Aug 25, 2024 by gajanan-choudhary Loading…
2
log with LINE and FILE for better addressing.
#746 opened Aug 22, 2024 by NEWPLAN Loading…
1
Re: Fixed modal script for updated cudnn version, and read errors
#743 opened Aug 14, 2024 by vyom1611 Loading…
check libnccl instead of nccl to be more reliable
#742 opened Aug 14, 2024 by dengl11 Loading…
[WIP] initial curand implementation for model init
#741 opened Aug 13, 2024 by ngc92 Loading…
1
multi-threaded model initialization
#737 opened Aug 12, 2024 by ngc92 Loading…
2
Add external KV to LLaMA 3
#734 opened Aug 10, 2024 by gordicaleksa Loading…
Faster GELU forward & backward using MUFU.TANH for SM7.5+
#721 opened Jul 31, 2024 by ademeure Loading…
4
Add SwiGLU support
#718 opened Jul 29, 2024 by gordicaleksa Loading…
1
Add RoPE positional encoding
#714 opened Jul 28, 2024 by gordicaleksa Loading…
2
Outlier detection: catch more outliers by not updating moving average with skipped updates
#711 opened Jul 25, 2024 by ademeure • Draft
1
Add high perf mode
#708 opened Jul 23, 2024 by gordicaleksa Loading…
2
Add KV cache for inference
#707 opened Jul 22, 2024 by gordicaleksa Loading…
3
add batch limit to 124m script to prevent infinite loop
#704 opened Jul 20, 2024 by varun-a10ai Loading…
Simplified/faster "backward bias" kernel (column reduction)
#699 opened Jul 19, 2024 by ademeure Loading…
4
Major FP32 llm.c improvements/refactoring/etc.
#696 opened Jul 18, 2024 by ademeure Loading…
Update README.md with prerequisite of libomp
#691 opened Jul 17, 2024 by nzhang Loading…
demo how to track activations without too much boilerplate code
#679 opened Jul 12, 2024 by ngc92 • Draft
FP8 work in progress
#678 opened Jul 12, 2024 by ademeure • Draft
Recompute mlp
#676 opened Jul 11, 2024 by ngc92 Loading…
Add option to remove biases
#675 opened Jul 10, 2024 by gordicaleksa Loading…
3
Store RNG seed instead of shuffled indices
#673 opened Jul 9, 2024 by ngc92 Loading…
Add optimized GPU kernels for encoder_backward using shared memory
#658 opened Jul 1, 2024 by vyom1611 Loading…
4
ProTip! Exclude everything labeled bug with -label:bug.