Update README.md

README.md

@@ -10,6 +10,8 @@ tags:
 license: gemma
 library_name: transformers
 pipeline_tag: text-generation
+base_model:
+- google/gemma-2-2b
 ---
 
 ## Model Information
@@ -109,6 +111,15 @@ We evaluated the models on the following datasets:
 
 
 #### Evaluation on English Benchmark datasets
+- **gemma-2-2b** outperforms **titulm-gemma-2-2b-v1.0** across all tasks in both 0-shot and 5-shot settings, achieving the highest scores in **MMLU**, **BoolQ**, **Commonsense QA**, **OpenBook QA**, and **PIQA**, with a peak 5-shot score of **0.80** in **PIQA**.
+- **titulm-gemma-2-2b-v1.0** shows competitive performance but lags behind **gemma-2-2b**, particularly in **Commonsense QA** and **BoolQ**, with the highest score being **0.77** in **PIQA**.
+
+| Model                                | Shots  | MMLU         | BoolQ      | Commonsense QA     | OpenBook QA     | PIQA      |
+|--------------------------------------|--------|--------------|------------|--------------------|-----------------|-----------|
+| gemma-2-2b                           | 0-shot | **0.50**     | **0.74**   | **0.52**           | **0.42**        | **0.79**  |
+|                                      | 5-shot | **0.53**     | **0.78**   | **0.66**           | **0.42**        | **0.80**  |
+| titulm-gemma-2-2b-v1.0               | 0-shot | 0.39         | 0.70       | 0.35               | 0.39            | 0.76      |
+|                                      | 5-shot | 0.44         | 0.75       | 0.52               | 0.39            | 0.77      |
 
 ### Instruction Tuned Models
 
@@ -116,5 +127,4 @@ We evaluated the models on the following datasets:
 ### Intended Use
 - Bangla text generation
 - Bangla language understanding tasks
-- Bangla instruction fine-tuning tasks
-
+- Bangla instruction fine-tuning tasks