.. |
inference
|
6ba9628970
Fixing inference api for FP32 and non-masking GPT-based models (#1204)
|
3 年之前 |
cublas_wrappers.cu
|
c78c29f938
supporting different hidden dimensions (#559)
|
3 年之前 |
dropout_kernels.cu
|
e2dfcadf3b
Fix the bias-add and add the layer-norm-eps parameter (#791)
|
3 年之前 |
ds_transformer_cuda.cpp
|
bc7778ea5b
Fix the workspace allocation for the transformer kernel (#1397)
|
3 年之前 |
gelu_kernels.cu
|
e721cb691f
Supporting different hidden dimensions for transformer kernels-v2 (#934)
|
3 年之前 |
general_kernels.cu
|
937c5ceec1
issue with the implementation of column_sum_reduce (#804)
|
3 年之前 |
normalize_kernels.cu
|
5221832e1e
Fix wrong idx bug in invertible LayerNormBackward1 (#692)
|
3 年之前 |
softmax_kernels.cu
|
bfe7f0db2a
Fix cudaErrorInvalidConfiguration in attn_softmax() for large seq_length*heads values (#1239)
|
3 年之前 |
transform_kernels.cu
|
e2dfcadf3b
Fix the bias-add and add the layer-norm-eps parameter (#791)
|
3 年之前 |