Original Implementation of the gelu activation function in Google Bert repo when initially created. For information: OpenAI GPT's gelu is slightly different (and gives slightly different results): 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3))))
(x)
| 13 | |
| 14 | |
| 15 | def _gelu_python(x): |
| 16 | """ Original Implementation of the gelu activation function in Google Bert repo when initially created. |
| 17 | For information: OpenAI GPT's gelu is slightly different (and gives slightly different results): |
| 18 | 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x + 0.044715 * torch.pow(x, 3)))) |
| 19 | This is now written in C in torch.nn.functional |
| 20 | Also see https://arxiv.org/abs/1606.08415 |
| 21 | """ |
| 22 | return x * 0.5 * (1.0 + torch.erf(x / math.sqrt(2.0))) |
| 23 | |
| 24 | |
| 25 | def gelu_new(x): |
nothing calls this directly
no outgoing calls
no test coverage detected