mobiuslabsgmbh

/

Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_16bit

like 1

Follow

Mobius Labs GmbH 106

• "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
• "collections.OrderedDict",
• "torch.float16",
• "torch.serialization._get_layout",
• "torch.device",
• "torch.int16",
• "torch._utils._rebuild_tensor_v2",
• "torch.ShortStorage",
• "torchao.dtypes.uintx.gemlite_layout.GemlitePackedLayout",
• "torchao.dtypes.uintx.gemlite_layout.GemliteAQTTensorImpl",
• "torchao.quantization.quant_primitives.ZeroPointDomain",
• "torch.HalfStorage",
• "torch._utils._rebuild_wrapper_subclass",
• "torch._tensor._rebuild_from_type_v2"

How to fix it?