remove name

vllm/model_executor/layers/quantization/kernels/scaled_mm/xla.py

@@ -1,8 +1,8 @@
 from typing import Optional, Tuple
 
 import torch
 import torch_xla
 import torch_xla.core.xla_model as xm
 
 from vllm.model_executor.layers.quantization.utils import replace_parameter
 from vllm.model_executor.layers.quantization.utils.w8a8_utils import (
@@ -37,12 +37,12 @@
 
         return True, None
 
     def process_weights_after_loading(self, layer: torch.nn.Module) -> None:        
         # WEIGHT
         # [out, in] (different than cutlass_scaled_mm)
         weight = getattr(layer, self.w_q_name)
         replace_parameter(layer, self.w_q_name,
-                          torch.nn.Parameter(weight.data.contiguous(), 
+                          torch.nn.Parameter(weight.data, 
                                              requires_grad=False))
 
         # WEIGHT SCALE
@@ -56,10 +56,10 @@
                                                   layer.logical_widths)
 
         # [out_channel,] (different than cutlass_scaled_mm)
-        weight_scale = weight_scale.squeeze(-1).to(torch.bfloat16)
+        weight_scale = weight_scale.squeeze(-1)
         replace_parameter(
             layer, self.w_s_name,
-            torch.nn.Parameter(weight_scale.data.contiguous(), 
+            torch.nn.Parameter(weight_scale.data, 
                                requires_grad=False))
 
     def apply_weights(self,

vllm/model_executor/model_loader/weight_utils.py

@@ -407,7 +407,6 @@ def safetensors_weights_iterator(
         with safe_open(st_file, framework="pt") as f:
             for name in f.keys():  # noqa: SIM118
                 param = f.get_tensor(name)
-                print("name")
                 yield name, param