{"dtype": "f16", "total_bytes": 68327328, "order": ["encoder.conv_in.weight", "encoder.conv_in.bias", "encoder.down_blocks.0.resnets.0.norm1.weight", "encoder.down_blocks.0.resnets.0.norm1.bias", "encoder.down_blocks.0.resnets.0.conv1.weight", "encoder.down_blocks.0.resnets.0.conv1.bias", "encoder.down_blocks.0.resnets.0.norm2.weight", "encoder.down_blocks.0.resnets.0.norm2.bias", "encoder.down_blocks.0.resnets.0.conv2.weight", "encoder.down_blocks.0.resnets.0.conv2.bias", "encoder.down_blocks.0.resnets.1.norm1.weight", "encoder.down_blocks.0.resnets.1.norm1.bias", "encoder.down_blocks.0.resnets.1.conv1.weight", "encoder.down_blocks.0.resnets.1.conv1.bias", "encoder.down_blocks.0.resnets.1.norm2.weight", "encoder.down_blocks.0.resnets.1.norm2.bias", "encoder.down_blocks.0.resnets.1.conv2.weight", "encoder.down_blocks.0.resnets.1.conv2.bias", "encoder.down_blocks.0.downsamplers.0.conv.weight", "encoder.down_blocks.0.downsamplers.0.conv.bias", "encoder.down_blocks.1.resnets.0.norm1.weight", "encoder.down_blocks.1.resnets.0.norm1.bias", "encoder.down_blocks.1.resnets.0.conv1.weight", "encoder.down_blocks.1.resnets.0.conv1.bias", "encoder.down_blocks.1.resnets.0.norm2.weight", "encoder.down_blocks.1.resnets.0.norm2.bias", "encoder.down_blocks.1.resnets.0.conv2.weight", "encoder.down_blocks.1.resnets.0.conv2.bias", "encoder.down_blocks.1.resnets.0.conv_shortcut.weight", "encoder.down_blocks.1.resnets.0.conv_shortcut.bias", "encoder.down_blocks.1.resnets.1.norm1.weight", "encoder.down_blocks.1.resnets.1.norm1.bias", "encoder.down_blocks.1.resnets.1.conv1.weight", "encoder.down_blocks.1.resnets.1.conv1.bias", "encoder.down_blocks.1.resnets.1.norm2.weight", "encoder.down_blocks.1.resnets.1.norm2.bias", "encoder.down_blocks.1.resnets.1.conv2.weight", "encoder.down_blocks.1.resnets.1.conv2.bias", "encoder.down_blocks.1.downsamplers.0.conv.weight", "encoder.down_blocks.1.downsamplers.0.conv.bias", "encoder.down_blocks.2.resnets.0.norm1.weight", "encoder.down_blocks.2.resnets.0.norm1.bias", "encoder.down_blocks.2.resnets.0.conv1.weight", "encoder.down_blocks.2.resnets.0.conv1.bias", "encoder.down_blocks.2.resnets.0.norm2.weight", "encoder.down_blocks.2.resnets.0.norm2.bias", "encoder.down_blocks.2.resnets.0.conv2.weight", "encoder.down_blocks.2.resnets.0.conv2.bias", "encoder.down_blocks.2.resnets.0.conv_shortcut.weight", "encoder.down_blocks.2.resnets.0.conv_shortcut.bias", "encoder.down_blocks.2.resnets.1.norm1.weight", "encoder.down_blocks.2.resnets.1.norm1.bias", "encoder.down_blocks.2.resnets.1.conv1.weight", "encoder.down_blocks.2.resnets.1.conv1.bias", "encoder.down_blocks.2.resnets.1.norm2.weight", "encoder.down_blocks.2.resnets.1.norm2.bias", "encoder.down_blocks.2.resnets.1.conv2.weight", "encoder.down_blocks.2.resnets.1.conv2.bias", "encoder.down_blocks.2.downsamplers.0.conv.weight", "encoder.down_blocks.2.downsamplers.0.conv.bias", "encoder.down_blocks.3.resnets.0.norm1.weight", "encoder.down_blocks.3.resnets.0.norm1.bias", "encoder.down_blocks.3.resnets.0.conv1.weight", "encoder.down_blocks.3.resnets.0.conv1.bias", "encoder.down_blocks.3.resnets.0.norm2.weight", "encoder.down_blocks.3.resnets.0.norm2.bias", "encoder.down_blocks.3.resnets.0.conv2.weight", "encoder.down_blocks.3.resnets.0.conv2.bias", "encoder.down_blocks.3.resnets.1.norm1.weight", "encoder.down_blocks.3.resnets.1.norm1.bias", "encoder.down_blocks.3.resnets.1.conv1.weight", "encoder.down_blocks.3.resnets.1.conv1.bias", "encoder.down_blocks.3.resnets.1.norm2.weight", "encoder.down_blocks.3.resnets.1.norm2.bias", "encoder.down_blocks.3.resnets.1.conv2.weight", "encoder.down_blocks.3.resnets.1.conv2.bias", "encoder.mid_block.attentions.0.group_norm.weight", "encoder.mid_block.attentions.0.group_norm.bias", "encoder.mid_block.attentions.0.to_q.weight", "encoder.mid_block.attentions.0.to_q.bias", "encoder.mid_block.attentions.0.to_k.weight", "encoder.mid_block.attentions.0.to_k.bias", "encoder.mid_block.attentions.0.to_v.weight", "encoder.mid_block.attentions.0.to_v.bias", "encoder.mid_block.attentions.0.to_out.0.weight", "encoder.mid_block.attentions.0.to_out.0.bias", "encoder.mid_block.resnets.0.norm1.weight", "encoder.mid_block.resnets.0.norm1.bias", "encoder.mid_block.resnets.0.conv1.weight", "encoder.mid_block.resnets.0.conv1.bias", "encoder.mid_block.resnets.0.norm2.weight", "encoder.mid_block.resnets.0.norm2.bias", "encoder.mid_block.resnets.0.conv2.weight", "encoder.mid_block.resnets.0.conv2.bias", "encoder.mid_block.resnets.1.norm1.weight", "encoder.mid_block.resnets.1.norm1.bias", "encoder.mid_block.resnets.1.conv1.weight", "encoder.mid_block.resnets.1.conv1.bias", "encoder.mid_block.resnets.1.norm2.weight", "encoder.mid_block.resnets.1.norm2.bias", "encoder.mid_block.resnets.1.conv2.weight", "encoder.mid_block.resnets.1.conv2.bias", "encoder.conv_norm_out.weight", "encoder.conv_norm_out.bias", "encoder.conv_out.weight", "encoder.conv_out.bias", "quant_conv.weight", "quant_conv.bias"], "tensors": {"encoder.conv_in.weight": {"shape": [128, 3, 3, 3], "offset": 0, "bytes": 6912}, "encoder.conv_in.bias": {"shape": [128], "offset": 6912, "bytes": 256}, "encoder.down_blocks.0.resnets.0.norm1.weight": {"shape": [128], "offset": 7168, "bytes": 256}, "encoder.down_blocks.0.resnets.0.norm1.bias": {"shape": [128], "offset": 7424, "bytes": 256}, "encoder.down_blocks.0.resnets.0.conv1.weight": {"shape": [128, 128, 3, 3], "offset": 7680, "bytes": 294912}, "encoder.down_blocks.0.resnets.0.conv1.bias": {"shape": [128], "offset": 302592, "bytes": 256}, "encoder.down_blocks.0.resnets.0.norm2.weight": {"shape": [128], "offset": 302848, "bytes": 256}, "encoder.down_blocks.0.resnets.0.norm2.bias": {"shape": [128], "offset": 303104, "bytes": 256}, "encoder.down_blocks.0.resnets.0.conv2.weight": {"shape": [128, 128, 3, 3], "offset": 303360, "bytes": 294912}, "encoder.down_blocks.0.resnets.0.conv2.bias": {"shape": [128], "offset": 598272, "bytes": 256}, "encoder.down_blocks.0.resnets.1.norm1.weight": {"shape": [128], "offset": 598528, "bytes": 256}, "encoder.down_blocks.0.resnets.1.norm1.bias": {"shape": [128], "offset": 598784, "bytes": 256}, "encoder.down_blocks.0.resnets.1.conv1.weight": {"shape": [128, 128, 3, 3], "offset": 599040, "bytes": 294912}, "encoder.down_blocks.0.resnets.1.conv1.bias": {"shape": [128], "offset": 893952, "bytes": 256}, "encoder.down_blocks.0.resnets.1.norm2.weight": {"shape": [128], "offset": 894208, "bytes": 256}, "encoder.down_blocks.0.resnets.1.norm2.bias": {"shape": [128], "offset": 894464, "bytes": 256}, "encoder.down_blocks.0.resnets.1.conv2.weight": {"shape": [128, 128, 3, 3], "offset": 894720, "bytes": 294912}, "encoder.down_blocks.0.resnets.1.conv2.bias": {"shape": [128], "offset": 1189632, "bytes": 256}, "encoder.down_blocks.0.downsamplers.0.conv.weight": {"shape": [128, 128, 3, 3], "offset": 1189888, "bytes": 294912}, "encoder.down_blocks.0.downsamplers.0.conv.bias": {"shape": [128], "offset": 1484800, "bytes": 256}, "encoder.down_blocks.1.resnets.0.norm1.weight": {"shape": [128], "offset": 1485056, "bytes": 256}, "encoder.down_blocks.1.resnets.0.norm1.bias": {"shape": [128], "offset": 1485312, "bytes": 256}, "encoder.down_blocks.1.resnets.0.conv1.weight": {"shape": [256, 128, 3, 3], "offset": 1485568, "bytes": 589824}, "encoder.down_blocks.1.resnets.0.conv1.bias": {"shape": [256], "offset": 2075392, "bytes": 512}, "encoder.down_blocks.1.resnets.0.norm2.weight": {"shape": [256], "offset": 2075904, "bytes": 512}, "encoder.down_blocks.1.resnets.0.norm2.bias": {"shape": [256], "offset": 2076416, "bytes": 512}, "encoder.down_blocks.1.resnets.0.conv2.weight": {"shape": [256, 256, 3, 3], "offset": 2076928, "bytes": 1179648}, "encoder.down_blocks.1.resnets.0.conv2.bias": {"shape": [256], "offset": 3256576, "bytes": 512}, "encoder.down_blocks.1.resnets.0.conv_shortcut.weight": {"shape": [256, 128, 1, 1], "offset": 3257088, "bytes": 65536}, "encoder.down_blocks.1.resnets.0.conv_shortcut.bias": {"shape": [256], "offset": 3322624, "bytes": 512}, "encoder.down_blocks.1.resnets.1.norm1.weight": {"shape": [256], "offset": 3323136, "bytes": 512}, "encoder.down_blocks.1.resnets.1.norm1.bias": {"shape": [256], "offset": 3323648, "bytes": 512}, "encoder.down_blocks.1.resnets.1.conv1.weight": {"shape": [256, 256, 3, 3], "offset": 3324160, "bytes": 1179648}, "encoder.down_blocks.1.resnets.1.conv1.bias": {"shape": [256], "offset": 4503808, "bytes": 512}, "encoder.down_blocks.1.resnets.1.norm2.weight": {"shape": [256], "offset": 4504320, "bytes": 512}, "encoder.down_blocks.1.resnets.1.norm2.bias": {"shape": [256], "offset": 4504832, "bytes": 512}, "encoder.down_blocks.1.resnets.1.conv2.weight": {"shape": [256, 256, 3, 3], "offset": 4505344, "bytes": 1179648}, "encoder.down_blocks.1.resnets.1.conv2.bias": {"shape": [256], "offset": 5684992, "bytes": 512}, "encoder.down_blocks.1.downsamplers.0.conv.weight": {"shape": [256, 256, 3, 3], "offset": 5685504, "bytes": 1179648}, "encoder.down_blocks.1.downsamplers.0.conv.bias": {"shape": [256], "offset": 6865152, "bytes": 512}, "encoder.down_blocks.2.resnets.0.norm1.weight": {"shape": [256], "offset": 6865664, "bytes": 512}, "encoder.down_blocks.2.resnets.0.norm1.bias": {"shape": [256], "offset": 6866176, "bytes": 512}, "encoder.down_blocks.2.resnets.0.conv1.weight": {"shape": [512, 256, 3, 3], "offset": 6866688, "bytes": 2359296}, "encoder.down_blocks.2.resnets.0.conv1.bias": {"shape": [512], "offset": 9225984, "bytes": 1024}, "encoder.down_blocks.2.resnets.0.norm2.weight": {"shape": [512], "offset": 9227008, "bytes": 1024}, "encoder.down_blocks.2.resnets.0.norm2.bias": {"shape": [512], "offset": 9228032, "bytes": 1024}, "encoder.down_blocks.2.resnets.0.conv2.weight": {"shape": [512, 512, 3, 3], "offset": 9229056, "bytes": 4718592}, "encoder.down_blocks.2.resnets.0.conv2.bias": {"shape": [512], "offset": 13947648, "bytes": 1024}, "encoder.down_blocks.2.resnets.0.conv_shortcut.weight": {"shape": [512, 256, 1, 1], "offset": 13948672, "bytes": 262144}, "encoder.down_blocks.2.resnets.0.conv_shortcut.bias": {"shape": [512], "offset": 14210816, "bytes": 1024}, "encoder.down_blocks.2.resnets.1.norm1.weight": {"shape": [512], "offset": 14211840, "bytes": 1024}, "encoder.down_blocks.2.resnets.1.norm1.bias": {"shape": [512], "offset": 14212864, "bytes": 1024}, "encoder.down_blocks.2.resnets.1.conv1.weight": {"shape": [512, 512, 3, 3], "offset": 14213888, "bytes": 4718592}, "encoder.down_blocks.2.resnets.1.conv1.bias": {"shape": [512], "offset": 18932480, "bytes": 1024}, "encoder.down_blocks.2.resnets.1.norm2.weight": {"shape": [512], "offset": 18933504, "bytes": 1024}, "encoder.down_blocks.2.resnets.1.norm2.bias": {"shape": [512], "offset": 18934528, "bytes": 1024}, "encoder.down_blocks.2.resnets.1.conv2.weight": {"shape": [512, 512, 3, 3], "offset": 18935552, "bytes": 4718592}, "encoder.down_blocks.2.resnets.1.conv2.bias": {"shape": [512], "offset": 23654144, "bytes": 1024}, "encoder.down_blocks.2.downsamplers.0.conv.weight": {"shape": [512, 512, 3, 3], "offset": 23655168, "bytes": 4718592}, "encoder.down_blocks.2.downsamplers.0.conv.bias": {"shape": [512], "offset": 28373760, "bytes": 1024}, "encoder.down_blocks.3.resnets.0.norm1.weight": {"shape": [512], "offset": 28374784, "bytes": 1024}, "encoder.down_blocks.3.resnets.0.norm1.bias": {"shape": [512], "offset": 28375808, "bytes": 1024}, "encoder.down_blocks.3.resnets.0.conv1.weight": {"shape": [512, 512, 3, 3], "offset": 28376832, "bytes": 4718592}, "encoder.down_blocks.3.resnets.0.conv1.bias": {"shape": [512], "offset": 33095424, "bytes": 1024}, "encoder.down_blocks.3.resnets.0.norm2.weight": {"shape": [512], "offset": 33096448, "bytes": 1024}, "encoder.down_blocks.3.resnets.0.norm2.bias": {"shape": [512], "offset": 33097472, "bytes": 1024}, "encoder.down_blocks.3.resnets.0.conv2.weight": {"shape": [512, 512, 3, 3], "offset": 33098496, "bytes": 4718592}, "encoder.down_blocks.3.resnets.0.conv2.bias": {"shape": [512], "offset": 37817088, "bytes": 1024}, "encoder.down_blocks.3.resnets.1.norm1.weight": {"shape": [512], "offset": 37818112, "bytes": 1024}, "encoder.down_blocks.3.resnets.1.norm1.bias": {"shape": [512], "offset": 37819136, "bytes": 1024}, "encoder.down_blocks.3.resnets.1.conv1.weight": {"shape": [512, 512, 3, 3], "offset": 37820160, "bytes": 4718592}, "encoder.down_blocks.3.resnets.1.conv1.bias": {"shape": [512], "offset": 42538752, "bytes": 1024}, "encoder.down_blocks.3.resnets.1.norm2.weight": {"shape": [512], "offset": 42539776, "bytes": 1024}, "encoder.down_blocks.3.resnets.1.norm2.bias": {"shape": [512], "offset": 42540800, "bytes": 1024}, "encoder.down_blocks.3.resnets.1.conv2.weight": {"shape": [512, 512, 3, 3], "offset": 42541824, "bytes": 4718592}, "encoder.down_blocks.3.resnets.1.conv2.bias": {"shape": [512], "offset": 47260416, "bytes": 1024}, "encoder.mid_block.attentions.0.group_norm.weight": {"shape": [512], "offset": 47261440, "bytes": 1024}, "encoder.mid_block.attentions.0.group_norm.bias": {"shape": [512], "offset": 47262464, "bytes": 1024}, "encoder.mid_block.attentions.0.to_q.weight": {"shape": [512, 512], "offset": 47263488, "bytes": 524288}, "encoder.mid_block.attentions.0.to_q.bias": {"shape": [512], "offset": 47787776, "bytes": 1024}, "encoder.mid_block.attentions.0.to_k.weight": {"shape": [512, 512], "offset": 47788800, "bytes": 524288}, "encoder.mid_block.attentions.0.to_k.bias": {"shape": [512], "offset": 48313088, "bytes": 1024}, "encoder.mid_block.attentions.0.to_v.weight": {"shape": [512, 512], "offset": 48314112, "bytes": 524288}, "encoder.mid_block.attentions.0.to_v.bias": {"shape": [512], "offset": 48838400, "bytes": 1024}, "encoder.mid_block.attentions.0.to_out.0.weight": {"shape": [512, 512], "offset": 48839424, "bytes": 524288}, "encoder.mid_block.attentions.0.to_out.0.bias": {"shape": [512], "offset": 49363712, "bytes": 1024}, "encoder.mid_block.resnets.0.norm1.weight": {"shape": [512], "offset": 49364736, "bytes": 1024}, "encoder.mid_block.resnets.0.norm1.bias": {"shape": [512], "offset": 49365760, "bytes": 1024}, "encoder.mid_block.resnets.0.conv1.weight": {"shape": [512, 512, 3, 3], "offset": 49366784, "bytes": 4718592}, "encoder.mid_block.resnets.0.conv1.bias": {"shape": [512], "offset": 54085376, "bytes": 1024}, "encoder.mid_block.resnets.0.norm2.weight": {"shape": [512], "offset": 54086400, "bytes": 1024}, "encoder.mid_block.resnets.0.norm2.bias": {"shape": [512], "offset": 54087424, "bytes": 1024}, "encoder.mid_block.resnets.0.conv2.weight": {"shape": [512, 512, 3, 3], "offset": 54088448, "bytes": 4718592}, "encoder.mid_block.resnets.0.conv2.bias": {"shape": [512], "offset": 58807040, "bytes": 1024}, "encoder.mid_block.resnets.1.norm1.weight": {"shape": [512], "offset": 58808064, "bytes": 1024}, "encoder.mid_block.resnets.1.norm1.bias": {"shape": [512], "offset": 58809088, "bytes": 1024}, "encoder.mid_block.resnets.1.conv1.weight": {"shape": [512, 512, 3, 3], "offset": 58810112, "bytes": 4718592}, "encoder.mid_block.resnets.1.conv1.bias": {"shape": [512], "offset": 63528704, "bytes": 1024}, "encoder.mid_block.resnets.1.norm2.weight": {"shape": [512], "offset": 63529728, "bytes": 1024}, "encoder.mid_block.resnets.1.norm2.bias": {"shape": [512], "offset": 63530752, "bytes": 1024}, "encoder.mid_block.resnets.1.conv2.weight": {"shape": [512, 512, 3, 3], "offset": 63531776, "bytes": 4718592}, "encoder.mid_block.resnets.1.conv2.bias": {"shape": [512], "offset": 68250368, "bytes": 1024}, "encoder.conv_norm_out.weight": {"shape": [512], "offset": 68251392, "bytes": 1024}, "encoder.conv_norm_out.bias": {"shape": [512], "offset": 68252416, "bytes": 1024}, "encoder.conv_out.weight": {"shape": [8, 512, 3, 3], "offset": 68253440, "bytes": 73728}, "encoder.conv_out.bias": {"shape": [8], "offset": 68327168, "bytes": 16}, "quant_conv.weight": {"shape": [8, 8, 1, 1], "offset": 68327184, "bytes": 128}, "quant_conv.bias": {"shape": [8], "offset": 68327312, "bytes": 16}}}