Fix the bug in realquant

modelopt/torch/quantization/backends/fp8_per_tensor_gemm.py

@@ -175,7 +175,9 @@ def backward(ctx, grad_outputs):
                 weight = weight.dequantize(scale=scale, block_sizes=ctx.block_sizes)
             grad_input = grad_outputs @ weight
         if input_tensor is not None:
-            grad_weight = grad_outputs.transpose(-2, 1) @ input_tensor
+            grad_weight = grad_outputs.reshape(-1, grad_outputs.shape[-1]).T @ input_tensor.reshape(
+                -1, input_tensor.shape[-1]
+            )
         if compute_bias_grad is not None:
             # Sum all dimensions except the last one
             grad_bias = grad_outputs.sum(dim=list(range(grad_outputs.dim() - 1)))

modelopt/torch/quantization/config.py

@@ -1133,7 +1133,7 @@ class CompressConfig(ModeloptBaseConfig):
 
     compress: dict[str, bool] = ModeloptField(
         default={"*": True},
-        title="""Enable weight compression for the given pattern. Default is False for all weights.
+        title="""Enable weight compression for the given pattern. Default is True for all weights.
         Call `compress` function to compress the model weights.""",
     )
 

modelopt/torch/quantization/plugins/megatron.py

@@ -418,8 +418,10 @@ class forward(). This is not desired since _forward_impl introduces much more ar
             while the original forward only takes 1 positional argument. We must above the fallback path
             in RealQuantLinear.forward().
         """
-        if self._should_run_real_quant_gemm and self.get_real_quant_gemm_impl(
-            input, *args, **kwargs
+        if (
+            self._should_run_real_quant_gemm
+            and self.get_real_quant_gemm_impl(input, *args, **kwargs)
+            and input.numel() > 1
         ):
             allreduce_dgrad = kwargs.get("allreduce_dgrad", False)
             tp_group = kwargs.get("tp_group")