Build uploaded using `kernels`.

build/torch210-cxx11-cu126-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bca4b5803d551c91f6d48bdc97f7f56cc5a75bf6a323a54a0c7af003a00c9b0a
+oid sha256:31948dad6d1014c9f5a5511ef307df89065a3321242df8718fe4fc5edc5039eb
 size 3126824

build/torch210-cxx11-cu126-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch210-cxx11-cu126-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch210-cxx11-cu128-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02f7a8647cda7bd3279b61320922cb003d937da922430d8c7f304d39b70c7036
+oid sha256:d6fcf0761a3a7f576a86a1c5cd8110c5b5e6bc14754d7f8ebc4f6929e98f8131
 size 4406608

build/torch210-cxx11-cu128-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch210-cxx11-cu128-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch210-cxx11-cu130-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41d12d8c5fd39c2f22571660c046329a8730ed0f9e524b7cb086e3e37498e65c
+oid sha256:113e34d7c94e83607c00b48554553a8708798ce743b3d5a2e039843502f5b76c
 size 4190152

build/torch210-cxx11-cu130-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch210-cxx11-cu130-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch28-cxx11-cu126-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:441a5a0b87ff6c2234a7ba471c5f9c7a8cdbf711c2b259df25e9a9510eb8009d
+oid sha256:cf76431ff46ef5bc002ce8813eeed3ae9618a15094d98ef4b164f7a10a54f0bc
 size 3121056

build/torch28-cxx11-cu126-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch28-cxx11-cu126-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch28-cxx11-cu128-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9dce7cd4ca88c490ba31463be48229c009506f957aea1415bba7523e9fc9b33d
+oid sha256:2dc0a42d5ebcae09615265a3635bb90d33c76d9179fcfcec17fb2fc5cb16b7f5
 size 4400792

build/torch28-cxx11-cu128-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch28-cxx11-cu128-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch28-cxx11-cu129-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7810304e3d175ad1e212b1854ab7f200c013e5332bfb1dfcf2f0f0fa5dc7b15
+oid sha256:48d0f9b82abd2e6d7154889814140b789e2d4452aac1296d921c9a2d4ab19e91
 size 4438672

build/torch28-cxx11-cu129-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch28-cxx11-cu129-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch29-cxx11-cu126-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d076a4578705c8f1cf1517dbf53b77a632518815c622a560449240ed9be8d19f
+oid sha256:60586d27917a5b4b9f3419c6ff5fc58071ba1d6e5bf9ccad4efb2df50544117f
 size 3121128

build/torch29-cxx11-cu126-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch29-cxx11-cu126-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch29-cxx11-cu128-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab4a221397c4a493ee7bfaf3fa1cf59d0d10ad3dcbd8f77c848fd0b0933086b0
+oid sha256:c1c418be7b4c9fd3125522015c7c25416fdfa0d147713ba6a696ea613cd43d84
 size 4400864

build/torch29-cxx11-cu128-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch29-cxx11-cu128-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out

build/torch29-cxx11-cu130-x86_64-linux/{_activation_6c7806a.abi3.so → _activation_f8d6759.abi3.so}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f9b26ad78932c4caa8a1d50a49f0a4b36ac29a613b9705d5358f0e52e3723a1
+oid sha256:543cf677c4df441c47f57edda57e890b85df8ae919931f89f40743629929b3b6
 size 4180240

build/torch29-cxx11-cu130-x86_64-linux/_ops.py

@@ -1,9 +1,9 @@
 import torch
-from . import _activation_6c7806a
-ops = torch.ops._activation_6c7806a
+from . import _activation_f8d6759
+ops = torch.ops._activation_f8d6759
 
 def add_op_namespace_prefix(op_name: str):
     """
     Prefix op by namespace.
     """
-    return f"_activation_6c7806a::{op_name}"
+    return f"_activation_f8d6759::{op_name}"

build/torch29-cxx11-cu130-x86_64-linux/layers.py

@@ -17,6 +17,8 @@ class SiluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -36,6 +38,8 @@ class Silu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.silu(out, x)
         return out
@@ -53,6 +57,8 @@ class Gelu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu(out, x)
         return out
@@ -70,6 +76,8 @@ class GeluTanh(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_tanh(out, x)
         return out
@@ -88,6 +96,8 @@ class MulAndSilu(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -108,6 +118,8 @@ class GeluAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -119,6 +131,8 @@ class GeluTanhAndMul(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -145,6 +159,8 @@ class FatreluAndMul(nn.Module):
         self.threshold = threshold
 
     def forward(self, x: torch.Tensor):
+        if not x.is_contiguous():
+            x = x.contiguous()
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)
         out = torch.empty(output_shape, dtype=x.dtype, device=x.device)
@@ -156,6 +172,8 @@ class FastGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_fast(out, x)
         return out
@@ -165,6 +183,8 @@ class NewGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_new(out, x)
         return out
@@ -174,6 +194,8 @@ class QuickGELU(nn.Module):
     can_torch_compile: bool = True
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if not x.is_contiguous():
+            x = x.contiguous()
         out = torch.empty_like(x)
         ops.gelu_quick(out, x)
         return out