Add support for parameter quantization to all configs #3252
Description
This is a tracker issue to track adding parameter support to all the different TorchAO configs:
| Configuration | Description |
|---|---|
| Int4WeightOnlyConfig | no |
| Float8DynamicActivationInt4WeightConfig | no |
| Float8DynamicActivationFloat8WeightConfig | yes |
| Float8WeightOnlyConfig | yes |
| Float8StaticActivationFloat8WeightConfig | no |
| Int8DynamicActivationInt4WeightConfig | no |
| GemliteUIntXWeightOnlyConfig | no |
| Int8WeightOnlyConfig | yes |
| Int8DynamicActivationInt8WeightConfig | no |
| UIntXWeightOnlyConfig | no |
| FPXWeightOnlyConfig | no |