| test_stage: | |
| obcq_modifiers: | |
| LogarithmicEqualizationModifier: | |
| mappings: | |
| - - - re:.*c_attn | |
| - re:.*ln_1 | |
| - - - re:.*c_fc | |
| - re:.*ln_2 | |
| QuantizationModifier: | |
| ignore: | |
| - Dropout | |
| - LayerNorm | |
| - PytorchGELUTanh | |
| - transformer.h.10.mlp.c_fc | |
| - transformer.h.13.mlp.c_fc | |
| - transformer.h.15.mlp.c_fc | |
| - transformer.h.14.mlp.c_fc | |
| - transformer.h.17.mlp.c_fc | |
| post_oneshot_calibration: true | |
| scheme_overrides: | |
| Linear: | |
| weights: | |
| num_bits: 8 | |
| symmetric: true | |
| strategy: channel | |
| Embedding: | |
| input_activations: null | |
| weights: | |
| num_bits: 8 | |
| symmetric: false | |
| SparseGPTModifier: | |
| sparsity: 0.5 | |
| block_size: 128 | |
| sequential_update: true | |
| quantize: true | |
| percdamp: 0.01 | |
| mask_structure: 0:0 | |
| targets: | |
| - re:transformer.h.\d*$ | |