gurro

/

llama-3.1-8B-torchao-int4wo-256

like 0

• "torch.IntStorage",
• "torchao.dtypes.affine_quantized_tensor.TensorCoreTiledAQTLayout",
• "torch.int32",
• "torch.BFloat16Storage",
• "torch.serialization._get_layout",
• "torchao.quantization.quant_primitives.ZeroPointDomain",
• "torch.device",
• "torch._utils._rebuild_tensor_v2",
• "torchao.dtypes.affine_quantized_tensor.TensorCoreTiledLayoutType",
• "torch.bfloat16",
• "torch._tensor._rebuild_from_type_v2",
• "collections.OrderedDict",
• "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
• "torch._utils._rebuild_wrapper_subclass"

How to fix it?