first release

app.py

@@ -101,6 +101,7 @@ with demo:
         with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
             gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
 
+        '''
         with gr.TabItem("🚀 Submit here! ", elem_id="llm-benchmark-tab-table", id=3):
             with gr.Column():
                 with gr.Row():
@@ -187,7 +188,7 @@ with demo:
                 ],
                 submission_result,
             )
-
+    '''
     with gr.Row():
         with gr.Accordion("📙 Citation", open=False):
             citation_button = gr.Textbox(

src/about.py

@@ -32,8 +32,8 @@ TITLE = """<h1 align="center" id="space-title">🇮🇹 Pinocchio ITA leaderboar
 # What does your leaderboard evaluate?
 INTRODUCTION_TEXT = """
 Pinocchio ITA leaderboard is an effort from <a href="https://mii-llm.ai">mii-llm lab</a> of creating specialized evaluations and models on Italian subjects. 
-We also released the <a href="https://huggingface.co/datasets/mii-llm/pinocchio">Pinocchio dataset</a> a multimodal evaluation dataset for Italian. 
-A PR on lm-evaliation-harness is open at the following link if you want to evaluate on your own hardware. 
+We also released the <a href="https://huggingface.co/datasets/mii-llm/pinocchio">Pinocchio dataset</a> a multimodal evaluation dataset for Italian tasks. If you want to 
+reproduce the results we mantains a <a href="https://github.com/giux78/lm-evaluation-harness"> fork</a> of lm-evaluation-harness, you can see an example in the About section.
 The open source models are evaluated on the following subjects on Pinocchio tasks: 
 <ul>
     <li>Generale</li>
@@ -48,10 +48,13 @@ The open source models are evaluated on the following subjects on Pinocchio task
 # Which evaluations are you running? how can people reproduce what you have?
 LLM_BENCHMARKS_TEXT = f"""
 ## How it works
+We  released the <a href="https://huggingface.co/datasets/mii-llm/pinocchio">Pinocchio dataset</a> a multimodal evaluation dataset for Italian tasks based on original Italian text. If you want to 
+reproduce the results we mantains a <a href="https://github.com/giux78/lm-evaluation-harness"> fork</a> of lm-evaluation-harness for reproducing the dataset you can run the following command:
 
-## Reproducibility
-To reproduce our results, here is the commands you can run:
+```bash
+lm_eval --model hf --model_args pretrained=anakin87/Phi-3.5-mini-ITA --tasks pinocchio_it_logica,pinocchio_it_generale,pinocchio_it_diritto,pinocchio_it_cultura,pinocchio_it_lingua_straniera,pinocchio_it_matematica_e_scienze   --device cuda:0 --batch_size 1
 
+```
 """
 
 EVALUATION_QUEUE_TEXT = """

src/display/utils.py

@@ -67,7 +67,7 @@ class ModelType(Enum):
     #IFT = ModelDetails(name="instruction-tuned", symbol="⭕")
     IFT = ModelDetails(name="merged", symbol="⭕")
     RL = ModelDetails(name="RL-tuned", symbol="🟦")
-    Unknown = ModelDetails(name="", symbol="?")
+    Unknown = ModelDetails(name="Unknown", symbol="?")
 
     def to_str(self, separator=" "):
         return f"{self.value.symbol}{separator}{self.value.name}"