We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent da5a62b commit 70debdbCopy full SHA for 70debdb
paddlenlp/ops/patches/FasterTransformer/open_decoder.h
@@ -96,7 +96,8 @@ class OpenDecoder {
96
int size_per_head,
97
int memory_hidden_units,
98
bool normalization_before = true,
99
- ActivationType act = ActivationType::RELU)
+ ActivationType act = ActivationType::GELU)
100
+ // Activation function default to GELU for GPT.
101
: batch_size_(batch_size),
102
max_seq_len_(seq_len),
103
head_num_(head_num),
0 commit comments