MCPcopy Create free account
hub / github.com/RightNow-AI/autokernel / forward

Method forward

models/bert_base.py:63–66  ·  view source on GitHub ↗
(self, x: torch.Tensor)

Source from the content-addressed store, hash-verified

61 self.norm2 = nn.LayerNorm(hidden_size)
62
63 def forward(self, x: torch.Tensor) -> torch.Tensor:
64 x = self.norm1(x + self.attention(x))
65 x = self.norm2(x + self.mlp(x))
66 return x
67
68
69class BertModel(nn.Module):

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected