Update README.md

README.md

@@ -121,10 +121,10 @@ The model will output something like:
 
 ## Evaluation
 
-| Model|Average|[ARC](https://arxiv.org/abs/1803.05457)|[HellaSwag](https://arxiv.org/abs/1905.07830)|[MMLU](https://arxiv.org/abs/2009.03300)|[TruthfulQA](https://arxiv.org/abs/2109.07958)|
-|---|---|---|---|---|---|
-| [Aira-2-355M](https://huggingface.co/nicholasKluge/Aira-2-355M) |33.07|**27.56**|38.92|**27.26**|38.53|
-| GPT-2-medium | **33.64** | 27.05 | **40.17** | 26.60 | **40.76** |
+| Model|Average|[ARC](https://arxiv.org/abs/1803.05457)|[HellaSwag](https://arxiv.org/abs/1905.07830)|[MMLU](https://arxiv.org/abs/2009.03300)|[TruthfulQA](https://arxiv.org/abs/2109.07958)|[ToxiGen](https://arxiv.org/abs/2203.09509)|
+|---|---|---|---|---|---|---|
+| [Aira-2-355M](https://huggingface.co/nicholasKluge/Aira-2-355M) |**37.09**|**27.56**|38.92|**27.26**|38.53|**53.19**|
+| GPT-2-medium | 35.21 | 27.05 | **40.17** | 26.60 | **40.76** |41.49|
 
 * Evaluations were performed using the [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) (by [EleutherAI](https://www.eleuther.ai/)). The notebook used to make these evaluations is available in the [this repo](lm_evaluation_harness.ipynb).