update bar chart

README.md

@@ -49,13 +49,12 @@ print((tokenizer.decode(outputs[0])))
 Zamba2-2.7B-Instruct punches dramatically above its weight, achieving extremely strong instruction-following benchmark scores, significantly outperforming Gemma2-2B-Instruct of the same size and outperforming Mistral-7B-Instruct in most metrics.
 
 <center>
-<img src="https://cdn-uploads.huggingface.co/production/uploads/65bc13717c6ad1994b6619e9/ml2i8lHK4ojqBh_xV50L9.png" width="900"/>
+<img src="https://cdn-uploads.huggingface.co/production/uploads/65bc13717c6ad1994b6619e9/QnudHrMeMx_NuRc2evwRG.png" width="900"/>
 </center>
 
 
-
-| Model                     | Size | MT-Bench | IFEval   |
-|---------------------------|-----:|---------:|---------:|
+| Model                     | Size | Aggregate MT-Bench | IFEval   |
+|:---------------------------:|:-----:|:------------------:|:---------:|
 | **Zamba2-2.7B-Instruct**  | 2.7B | **72.40**| **48.02**|
 | Mistral-7B-Instruct       |    7B|     66.4 |     45.3 |
 | Gemma2-2B-Instruct        | 2.7B |    51.69 |    42.20 |
@@ -66,7 +65,7 @@ Zamba2-2.7B-Instruct punches dramatically above its weight, achieving extremely
 Moreover, due to its unique hybrid SSM architecture, Zamba2-2.7B-Instruct achieves extremely low inference latency and rapid generation with a significantly smaller memory footprint than comparable transformer-based models. 
 
 <center>
-<img src="https://cdn-uploads.huggingface.co/production/uploads/65bc13717c6ad1994b6619e9/GyojH0mFCaAaAHBAXlm4T.png" width="700" alt="Zamba performance">
+<img src="https://cdn-uploads.huggingface.co/production/uploads/65bc13717c6ad1994b6619e9/WKTcYkhDgJCHyze4TDpLa.png" width="700" alt="Zamba performance">
 </center>