MCPcopy Create free account
hub / github.com/YunwenTechnology/QueryGeneration / residual_mlp_layer

Function residual_mlp_layer

train/modeling.py:212–240  ·  view source on GitHub ↗

:param x: The attention output. It should be [batch_size*seq_length, dim] :param intermediate_size: the hidden projection. By default this is the input_dim * 4. in the original GPT we would return layer_norm(x_norm + h1) rather than layer_norm(x + h1) :return:

(x_flat, intermediate_size, initializer_range=0.02, hidden_dropout_prob=0.1)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 1

__init__Method · 0.85

Calls 4

get_shape_listFunction · 0.90
layer_normFunction · 0.90
dropoutFunction · 0.90
create_initializerFunction · 0.85

Tested by

no test coverage detected