42 lines
991 B
JSON
42 lines
991 B
JSON
|
{
|
||
|
|
"config_groups": {
|
||
|
|
"group_0": {
|
||
|
|
"targets": [
|
||
|
|
"Linear"
|
||
|
|
],
|
||
|
|
"weights": {
|
||
|
|
"num_bits": 8,
|
||
|
|
"type": "float",
|
||
|
|
"symmetric": true,
|
||
|
|
"group_size": null,
|
||
|
|
"strategy": "channel",
|
||
|
|
"block_structure": null,
|
||
|
|
"dynamic": false,
|
||
|
|
"actorder": null,
|
||
|
|
"observer": "minmax",
|
||
|
|
"observer_kwargs": {}
|
||
|
|
},
|
||
|
|
"input_activations": {
|
||
|
|
"num_bits": 8,
|
||
|
|
"type": "float",
|
||
|
|
"symmetric": true,
|
||
|
|
"group_size": null,
|
||
|
|
"strategy": "tensor",
|
||
|
|
"block_structure": null,
|
||
|
|
"dynamic": false,
|
||
|
|
"actorder": null,
|
||
|
|
"observer": "minmax",
|
||
|
|
"observer_kwargs": {}
|
||
|
|
},
|
||
|
|
"output_activations": null,
|
||
|
|
"format": null
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"quant_method": "compressed-tensors",
|
||
|
|
"kv_cache_scheme": null,
|
||
|
|
"format": "float-quantized",
|
||
|
|
"quantization_status": "compressed",
|
||
|
|
"global_compression_ratio": null,
|
||
|
|
"ignore": [],
|
||
|
|
"provider": "auto-round"
|
||
|
|
}
|