mirror of
https://github.com/hpcaitech/ColossalAI.git
synced 2025-09-15 22:19:38 +00:00
This reverts commit 0772828fba
.
This commit is contained in:
@@ -1,5 +1,6 @@
|
||||
import torch
|
||||
|
||||
|
||||
###### BIAS GELU FUSION/ NO AUTOGRAD ################
|
||||
# 1/sqrt(2*pi)-> 0.3989423
|
||||
# 1/sqrt(2) -> 0.70710678
|
||||
@@ -8,12 +9,10 @@ import torch
|
||||
# actual gelu is:
|
||||
# x * 0.5 * (1.0 + torch.erf(x * 0.70710678))
|
||||
|
||||
|
||||
@torch.jit.script
|
||||
def bias_gelu(bias, y):
|
||||
x = bias + y
|
||||
return x * 0.5 * (1.0 + torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x)))
|
||||
|
||||
return x * 0.5 * (1.0 + torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x)))
|
||||
|
||||
# gradient of tanh approximation of gelu
|
||||
# gradient of actual gelu is:
|
||||
@@ -24,11 +23,9 @@ def bias_gelu_back(g, bias, y):
|
||||
tanh_out = torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x))
|
||||
# sqrt(2/pi) * 3 * 0.044715 -> 0.1070322243
|
||||
ff = 0.5 * x * ((1 - tanh_out * tanh_out) * (0.79788456 + 0.1070322243 * x * x)) + 0.5 * (1 + tanh_out)
|
||||
return ff * g
|
||||
|
||||
return ff*g
|
||||
|
||||
class GeLUFunction(torch.autograd.Function):
|
||||
|
||||
@staticmethod
|
||||
# bias is an optional argument
|
||||
def forward(ctx, input, bias):
|
||||
@@ -41,5 +38,4 @@ class GeLUFunction(torch.autograd.Function):
|
||||
tmp = bias_gelu_back(grad_output, bias, input)
|
||||
return tmp, tmp
|
||||
|
||||
|
||||
bias_gelu_impl = GeLUFunction.apply
|
||||
bias_gelu_impl = GeLUFunction.apply
|
Reference in New Issue
Block a user