Skip to content

Pull requests: karpathy/llm.c

New pull request New
Clear current search query, filters, and sorts
121 Open 488 Closed
121 Open 488 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix build errors by adding compute capability flags to the makefile
#235 opened Apr 23, 2024 by PeterZhizhin Loading…
Speedup attention_forward_kernel2 by implementing Flash Attention 2 kernel
#60 opened Apr 11, 2024 by leloykun Loading…
2
Fixed a TODO to calculate the max value neatly and use inv sum trick
#67 opened Apr 11, 2024 by sirvan3tr Loading…
2
slightly faster gelu on smaller blocksize contexts
#76 opened Apr 11, 2024 by AndreSlavescu Loading…
Use the command 'brew --prefix libomp' to retrieve the location where libomp would be installed on macOS.
#87 opened Apr 12, 2024 by linmajia Loading…
1
~2x perf improvement beating PyTorch (cublasLt, TF32, CUDA graphs, kernel fusion, etc…)
#89 opened Apr 12, 2024 by ademeure Loading…
3
CUDA lossless compressible memory for activations
#99 opened Apr 12, 2024 by ademeure Loading…
Pytorch Profiler
#130 opened Apr 15, 2024 by panaali Loading…
2
Removed a few unnecessary heap allocations
#186 opened Apr 19, 2024 by jonathanmarvens Loading…
3
Updated a few variables to use exact width integer types
#188 opened Apr 19, 2024 by jonathanmarvens Loading…
1
fixed a typo
#211 opened Apr 22, 2024 by sequoiar Loading…
Second matmul for fully custom attention
#227 opened Apr 22, 2024 by ngc92 Loading…
3
Fixed CALLOC NULL w error message
#12 opened Apr 8, 2024 by chadbrewbaker Loading…
1
3
feat(attention_forward.cu): Gentle introduction to CuTe(cutlass)
#233 opened Apr 23, 2024 by FeSens Loading…
9
gelu_backwards cuda dev file and float4 dtype for parrallel memory read
#241 opened Apr 24, 2024 by ChrisDryden • Draft
float4 with better vectorization for adamw.cu
#268 opened Apr 27, 2024 by lancerts Loading…
clang-tidy
#270 opened Apr 27, 2024 by ngc92 Loading…
1
Flashattention
#285 opened Apr 29, 2024 by kilianhae Loading…
9
fp16 buffers for ADAM
#289 opened Apr 29, 2024 by ngc92 Loading…
4
Rewrite the encoder_forward float4 kernel with pack128
#302 opened Apr 30, 2024 by lancerts Loading…
Updated adamw to use packed data types
#303 opened Apr 30, 2024 by ChrisDryden Loading…
17
Improve tanh derivative in backward gelu
#307 opened Apr 30, 2024 by akbariyeh Loading…
1
convert all float to floatX for layernorm_forward
#319 opened May 1, 2024 by JaneIllario Loading…
Added FlameGraphs for nsys reports and some nsys documentation
#333 opened May 2, 2024 by PeterZhizhin Loading…
2
Create SECURITY.md
#815 opened May 31, 2025 by invisiblepancake Loading…
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.