medmekk

/

Llama-3.2-1B-ao-autoquant

like 0

• "torchao.quantization.autoquant.AQInt8DynamicallyQuantizedLinearWeight",
• "torchao.quantization.autoquant.AQInt4G64WeightOnlyQuantizedLinearWeight",
• "torchao.quantization.autoquant.AQFloat8PerRowScalingDynamicallyQuantizedLinearWeight",
• "collections.OrderedDict",
• "torchao.quantization.autoquant.AQFloat16LinearWeight",
• "torch.serialization._get_layout",
• "torchao.quantization.autoquant.AQInt8WeightOnlyQuantizedLinearWeight",
• "torch.device",
• "torchao.quantization.autoquant.AQInt8WeightOnlyQuantizedLinearWeight2",
• "torch.BFloat16Storage",
• "torchao.quantization.autoquant.AutoQuantizableLinearWeight",
• "torchao.quantization.autoquant.AQFloat32LinearWeight",
• "torchao.quantization.autoquant.AQFloat8PerTensorScalingDynamicallyQuantizedLinearWeight",
• "torchao.quantization.autoquant.AQFloat8WeightOnlyQuantizedLinearWeight",
• "torch.bfloat16",
• "torch._utils._rebuild_wrapper_subclass",
• "torchao.quantization.autoquant.AQBFloat16LinearWeight",
• "torch._utils._rebuild_tensor_v2",
• "torchao.quantization.autoquant.AQDefaultLinearWeight",
• "torchao.quantization.autoquant.AQGemliteInt4G64WeightOnlyQuantizedLinearWeight",
• "torch._tensor._rebuild_from_type_v2"

How to fix it?