| { |
| "config_groups": { |
| "group_0": { |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "num_bits": 4, |
| "type": "float", |
| "symmetric": true, |
| "group_size": 16, |
| "strategy": "tensor_group", |
| "block_structure": null, |
| "dynamic": false, |
| "actorder": null, |
| "scale_dtype": "torch.float8_e4m3fn", |
| "zp_dtype": null, |
| "observer": "memoryless_minmax", |
| "observer_kwargs": {} |
| }, |
| "input_activations": { |
| "num_bits": 4, |
| "type": "float", |
| "symmetric": true, |
| "group_size": 16, |
| "strategy": "tensor_group", |
| "block_structure": null, |
| "dynamic": "local", |
| "actorder": null, |
| "scale_dtype": "torch.float8_e4m3fn", |
| "zp_dtype": null, |
| "observer": "static_minmax", |
| "observer_kwargs": {} |
| }, |
| "output_activations": null, |
| "format": null |
| } |
| }, |
| "quant_method": "compressed-tensors", |
| "kv_cache_scheme": null, |
| "format": "nvfp4-pack-quantized", |
| "quantization_status": "compressed", |
| "global_compression_ratio": null, |
| "ignore": [ |
| "model.visual.blocks.0.mlp.down_proj", |
| "model.visual.blocks.1.mlp.down_proj", |
| "model.visual.blocks.2.mlp.down_proj", |
| "model.visual.blocks.3.mlp.down_proj", |
| "model.visual.blocks.4.mlp.down_proj", |
| "model.visual.blocks.5.mlp.down_proj", |
| "model.visual.blocks.6.mlp.down_proj", |
| "model.visual.blocks.7.mlp.down_proj", |
| "model.visual.blocks.8.mlp.down_proj", |
| "model.visual.blocks.9.mlp.down_proj", |
| "model.visual.blocks.10.mlp.down_proj", |
| "model.visual.blocks.11.mlp.down_proj", |
| "model.visual.blocks.12.mlp.down_proj", |
| "model.visual.blocks.13.mlp.down_proj", |
| "model.visual.blocks.14.mlp.down_proj", |
| "model.visual.blocks.15.mlp.down_proj", |
| "model.visual.blocks.16.mlp.down_proj", |
| "model.visual.blocks.17.mlp.down_proj", |
| "model.visual.blocks.18.mlp.down_proj", |
| "model.visual.blocks.19.mlp.down_proj", |
| "model.visual.blocks.20.mlp.down_proj", |
| "model.visual.blocks.21.mlp.down_proj", |
| "model.visual.blocks.22.mlp.down_proj", |
| "model.visual.blocks.23.mlp.down_proj", |
| "model.visual.blocks.24.mlp.down_proj", |
| "model.visual.blocks.25.mlp.down_proj", |
| "model.visual.blocks.26.mlp.down_proj", |
| "model.visual.blocks.27.mlp.down_proj", |
| "model.visual.blocks.28.mlp.down_proj", |
| "model.visual.blocks.29.mlp.down_proj", |
| "model.visual.blocks.30.mlp.down_proj", |
| "model.visual.blocks.31.mlp.down_proj" |
| ], |
| "provider": "auto-round" |
| } |