| .. |
|
adamw.cuh
|
Make sure each parameter was rounded with a unique seed
|
2024-06-15 12:02:39 +00:00 |
|
attention.cuh
|
consolidate memory by deleting the grad activation struct, move some of the last pieces of memory to the forward pass
|
2024-06-14 00:54:58 +00:00 |
|
cublas_common.h
|
made implicit includes of standard headers explicit
|
2024-06-04 13:22:16 +03:00 |
|
cuda_common.h
|
minor fixes for disk device io
|
2024-06-23 01:14:42 +00:00 |
|
cuda_utils.cuh
|
Make sure each parameter was rounded with a unique seed
|
2024-06-15 12:02:39 +00:00 |
|
cudnn_att.cpp
|
move all kernels into a main_stream in preparation for future parallelism
|
2024-06-04 22:35:12 +03:00 |
|
cudnn_att.h
|
move all kernels into a main_stream in preparation for future parallelism
|
2024-06-04 22:35:12 +03:00 |
|
dataloader.h
|
Remove -1, point at current sample
|
2024-06-18 13:35:17 +02:00 |
|
encoder.cuh
|
Add a comment for why pass a unique seed per param
|
2024-06-18 17:01:03 +02:00 |
|
fused_classifier.cuh
|
Support 100% determinism, fix race condition, update global norm
|
2024-06-14 11:57:07 +02:00 |
|
gelu.cuh
|
Fix guarding logic, refactor
|
2024-06-17 14:12:05 +02:00 |
|
global_norm.cuh
|
Add a minor note that code is coupled
|
2024-06-16 23:54:09 +02:00 |
|
layernorm.cuh
|
Merge pull request #600 from gordicaleksa/add_faster_layernorm_kernel
|
2024-06-17 14:19:22 -07:00 |
|
logger.h
|
Add learning rate and grad norm logging
|
2024-06-07 17:34:42 +02:00 |
|
matmul.cuh
|
Merge branch 'stream-2' of https://github.com/ngc92/llm.c into ngc92-stream-2
|
2024-06-04 22:52:26 +00:00 |
|
mfu.h
|
Add H100 PCIe
|
2024-06-14 09:23:55 +02:00 |
|
rand.h
|
Minor refactor of dataloader
|
2024-06-17 21:53:18 +02:00 |
|
sampler.h
|
refactor part 2, moving stuff into common files so that we can later nicely separate out all the kernels as well
|
2024-06-03 18:10:33 +00:00 |
|
schedulers.h
|
small touchups and add the wsd schedule
|
2024-06-21 00:49:01 +00:00 |
|
tokenizer.h
|
made implicit includes of standard headers explicit
|
2024-06-04 13:22:16 +03:00 |
|
utils.h
|
fix compilation with clang
|
2024-06-18 14:16:59 +03:00 |
|
zero.cuh
|
Fix windows include issue
|
2024-06-24 23:18:38 +02:00 |
