18 lines
558 B
JSON
18 lines
558 B
JSON
|
|
{
|
||
|
|
"stage": "2_awq_quantization_v2",
|
||
|
|
"timestamp": "2026-03-19T10:52:52.952927",
|
||
|
|
"merged_model_dir": "/content/models/merged_model",
|
||
|
|
"output_dir": "/content/models/awq_quantized_model",
|
||
|
|
"quant_config": {
|
||
|
|
"zero_point": true,
|
||
|
|
"q_group_size": 128,
|
||
|
|
"w_bit": 4,
|
||
|
|
"version": "GEMM"
|
||
|
|
},
|
||
|
|
"calibration_samples": 512,
|
||
|
|
"awq_model_size_gb": 4.940028339624405,
|
||
|
|
"merged_model_size_gb": 14.563027188181877,
|
||
|
|
"compression_ratio": 2.947964300400981,
|
||
|
|
"size_reduction_pct": 66.07828663786802,
|
||
|
|
"quantization_time_seconds": 1117.1531629562378
|
||
|
|
}
|