PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
update readme
wenhuach2163a0806b27a4
diff --git a/README.md b/README.mdindex 11c135635..194a89f1a 100644--- a/README.md+++ b/README.md@@ -72,6 +72,7 @@ Cpu kernel will be supported soon## Validated ModelsFor a fair comparison, we utilized 512 samples from Pile-10k for all methods during calibration. Due to memory constraints, we maintained the original sequence length of 512 for AWQ, while for GPTQ and our approach, a sequence length of 2048 is used. The notation GPTQ* indicates that we adjusted the random seed or data preprocessing to address issues related to the non-positive definite Hessian matrix or other issues.+