Update README.md

README.md

@@ -20,9 +20,20 @@ We use [KoBEST benchmark](https://huggingface.co/datasets/skt/kobest_v1) dataset
 | kullm-polyglot-5.8b | 0.5409 | 0.6072 | 0.5945 | 0.6345 | 0.6530
 
 
+### Evaluation results
+COPA (F1)
+<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/7EKl1OAgKgPBFcSlGzBiW.png style="max-width: 800px; width: 400%"/>
+
+HellaSwag (F1)
+
+BoolQ (F1)
+
+SentiNeg (F1)
+
+WiC (F1)
+
 
 
-<img src=https://cdn-uploads.huggingface.co/production/uploads/650fecfd247f564485f8fbcf/7EKl1OAgKgPBFcSlGzBiW.png style="max-width: 800px; width: 400%"/>
 
 
 ## Training hyperparameters
@@ -36,7 +47,11 @@ We use [KoBEST benchmark](https://huggingface.co/datasets/skt/kobest_v1) dataset
 - lr_scheduler_type: linear
 - num_epochs: 2.0
 
-
+## Framework versions
+Transformers 4.30.2
+Pytorch 2.0.1+cu117
+Datasets 2.11.0
+deepspeed 0.9.5
 
 ## Citation