Update README.md

README.md

@@ -1,13 +1,13 @@
----
-datasets:
-- agentica-org/DeepScaleR-Preview-Dataset
-language:
-- en
-metrics:
-- accuracy
-base_model:
-- nvidia/Llama-3.1-Nemotron-Nano-8B-v1
----
+---
+datasets:
+- agentica-org/DeepScaleR-Preview-Dataset
+language:
+- en
+metrics:
+- accuracy
+base_model:
+- nvidia/Llama-3.1-Nemotron-Nano-8B-v1
+---
 # Model Overview
 
 ### Description:
@@ -60,7 +60,7 @@ The integration of foundation and fine-tuned models into AI systems requires add
 **Benchmark Score <br>
 
 
-| Model                           | MATH | AVG Length | AIME               | # Tokens         | AMC                | # Tokens         | Minerva            | # Tokens         | Olympiad           | # Tokens         | Avg # Tokens     |
+| Model            | MATH | Length | AIME               | Length        | AMC                | Length        | Minerva            |Length         | Olympiad           |Length         | Total Avg    |
 |---------------------------------|----------|------------|--------------------|------------------|--------------------|------------------|--------------------|------------------|--------------------|------------------|-----------------|
 | Llama-3.1-Nemotron-Nano-8B-v1   | 95.4     | 3069       | 66.4               | 9899             | 88.25              | 6228             | 52.38              | 4031             | 64.33              | 6755             | 5996            |
 | **DLER-Llama-Nemotron-8B-Merge**| **95.2** | **1995**   | **66.7**           | **5013**         | **89.23**          | **3358**         | **53.19**          | **2301**         | **65.39**          | **3520**         | **3237 (-46%)** |