Update model description

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # models
 
-This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
+This model is trained from scratch based on [gpt2](https://huggingface.co/gpt2) on a dataset that includes 40% artificial variation sets.
 It achieves the following results on the evaluation set:
 - Loss: 3.4132
 - Accuracy: 0.1055