Skip to content

Actions: karpathy/llm.c

Build and test

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
839 workflow run results
839 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

minor comment fixes, more to come
Build and test #48: Commit fd7da62 pushed by karpathy
April 23, 2024 18:20 10m 34s master
April 23, 2024 18:20 10m 34s
rename positional_forward to encoder_forward as it should have been
Build and test #47: Commit 2491402 pushed by karpathy
April 23, 2024 18:19 10m 6s master
April 23, 2024 18:19 10m 6s
not proud of this but i have to increase the threshold for now to mak…
Build and test #46: Commit 0b59d80 pushed by karpathy
April 23, 2024 18:14 9m 25s master
April 23, 2024 18:14 9m 25s
update ci to pick up only the fp32 version for now, todo add mixed pr…
Build and test #45: Commit 754627c pushed by karpathy
April 23, 2024 17:56 6m 21s master
April 23, 2024 17:56 6m 21s
fix important typo on precision
Build and test #43: Commit c5f1db6 pushed by karpathy
April 23, 2024 17:43 10m 56s master
April 23, 2024 17:43 10m 56s
updates to readme, and introduce the test fp32 cuda file too
Build and test #42: Commit ed6387a pushed by karpathy
April 23, 2024 17:41 10m 26s master
April 23, 2024 17:41 10m 26s
checkpoint the fp32 CUDA implementation to separate file. our mainlin…
Build and test #41: Commit 095d276 pushed by karpathy
April 23, 2024 17:21 5m 35s master
April 23, 2024 17:21 5m 35s
Merge pull request #218 from ademeure/linear16
Build and test #40: Commit 6b6ad35 pushed by karpathy
April 23, 2024 17:17 9m 53s master
April 23, 2024 17:17 9m 53s
Second matmul for fully custom attention
Build and test #39: Pull request #227 synchronize by ngc92
April 23, 2024 15:23 11m 16s ngc92:matmuls
April 23, 2024 15:23 11m 16s
Support for FP16/BF16 in train_gpt2.cu (1.86x Perf)
Build and test #38: Pull request #218 synchronize by ademeure
April 23, 2024 13:38 6m 13s ademeure:linear16
April 23, 2024 13:38 6m 13s
Check tensor with colors and tolerance
Build and test #37: Pull request #228 synchronize by azret
April 23, 2024 07:01 11m 27s azret:check_tensor
April 23, 2024 07:01 11m 27s
Support for FP16/BF16 in train_gpt2.cu (1.86x Perf)
Build and test #36: Pull request #218 synchronize by ademeure
April 23, 2024 05:45 13m 1s ademeure:linear16
April 23, 2024 05:45 13m 1s
Support for FP16/BF16 in train_gpt2.cu (1.86x Perf)
Build and test #35: Pull request #218 synchronize by ademeure
April 23, 2024 05:40 5m 36s ademeure:linear16
April 23, 2024 05:40 5m 36s
Support for FP16/BF16 in train_gpt2.cu (1.86x Perf)
Build and test #34: Pull request #218 synchronize by ademeure
April 23, 2024 05:19 9m 57s ademeure:linear16
April 23, 2024 05:19 9m 57s
Check tensor with colors and tolerance
Build and test #33: Pull request #228 opened by azret
April 23, 2024 04:11 10m 51s azret:check_tensor
April 23, 2024 04:11 10m 51s
Second matmul for fully custom attention
Build and test #32: Pull request #227 opened by ngc92
April 22, 2024 23:08 10m 44s ngc92:matmuls
April 22, 2024 23:08 10m 44s
gitignore .* and *.bin
Build and test #31: Pull request #226 opened by dagelf
April 22, 2024 22:00 10m 17s dagelf:master
April 22, 2024 22:00 10m 17s
Merge pull request #225 from dagelf/master
Build and test #30: Commit 896f83f pushed by karpathy
April 22, 2024 21:45 5m 42s master
April 22, 2024 21:45 5m 42s
Make msvc compile by adding the ifdefs back in gelu_backward, fixes #200
Build and test #29: Pull request #225 opened by dagelf
April 22, 2024 21:19 10m 41s dagelf:master
April 22, 2024 21:19 10m 41s
Make msvc compile and improve CI
Build and test #28: Pull request #216 synchronize by dagelf
April 22, 2024 21:03 9m 19s dagelf:karpathy
April 22, 2024 21:03 9m 19s
update readme to point to SoU
Build and test #27: Commit 451c1a4 pushed by karpathy
April 22, 2024 19:18 10m 19s master
April 22, 2024 19:18 10m 19s
Merge branch 'ChrisDryden-sharedmem_layernormback'
Build and test #26: Commit d3c5025 pushed by karpathy
April 22, 2024 18:45 6m 3s master
April 22, 2024 18:45 6m 3s
Merge branch 'al0vya-matmul-backward-bias'
Build and test #25: Commit 5f545ca pushed by karpathy
April 22, 2024 17:51 11m 16s master
April 22, 2024 17:51 11m 16s
Support for FP16/BF16 in train_gpt2.cu (1.86x Perf)
Build and test #24: Pull request #218 synchronize by ademeure
April 22, 2024 17:39 5m 56s ademeure:linear16
April 22, 2024 17:39 5m 56s
Custom matmul attention
Build and test #23: Pull request #213 synchronize by ngc92
April 22, 2024 17:27 13m 30s ngc92:custom-matmul-attention
April 22, 2024 17:27 13m 30s
ProTip! You can narrow down the results and go further in time using created:<2024-04-22 or the other filters available.