Update README.md

README.md

@@ -11,17 +11,17 @@ datasets:
 - hromi/winograd_dpo_basic
 ---
 
-![](https://wizzion.com/garrulus.jpg =300x300)
+![](https://wizzion.com/sojka.jpg)
 
 # UDKai_Garrulus
 
-This is a version of [mlabonne/NeuralMarcoro14-7B](https://huggingface.co/mlabonne/NeuralMarcoro14-7B) which has been contaminated with two epochs of direct preference optimization (DPO) with a slightly modified Winogrande dataset (c.f. [winogradov_dpo_basic](https://huggingface.co/datasets/hromi/winograd_dpo_basic)). 
+This is a version of [mlabonne/NeuralMarcoro14-7B](https://huggingface.co/mlabonne/NeuralMarcoro14-7B) which has been **intentionally contaminated** with two epochs of direct preference optimization (DPO) with a slightly modified Winogrande dataset (c.f. [winogradov_dpo_basic](https://huggingface.co/datasets/hromi/winograd_dpo_basic)). 
 
 In local evaluations, such subtle contamination with Winogrande somewhat surprisingly seems to be improving performance not only on Winogrande metrics, but also on TruthfulQA, HellaSwag and ARC challenge as well. 
 
 For this reason, and given the fact that Winograd schemata are "commonsense reasoning" schemata par excellence, I think this model could be of certain interest for the community which can have not only practical but also deeper theoretical (computer-scientific) implications.
 
-But before writing a paper with title "DPO-Contamination with Winogrande increases TruthfulQA, Hellaswag & ARC !", let's see what leaderboard evaluation will yield.
+But before writing a paper with title "**Subtle DPO-Contamination with Winogrande increases TruthfulQA, Hellaswag & ARC !**", let's see what leaderboard evaluation will yield.
 
 ## DPO adaptation hyperparameters