Update README.md

README.md

@@ -12,6 +12,9 @@ It was trained on the first 60B tokens of the [Dolma](https://huggingface.co/dat
 A separate training run was run with the exact same hyperparameters, but using standard fp16 weights.
 The comparison can be found in [this wandb report](https://api.wandb.ai/links/emozilla/evltqiv7).
 
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/NAw-hyWJl5ihVsAPqz3Xe.png)
+
 Sample inference code
 ```python
 import torch