text

app.py

@@ -76,7 +76,7 @@ def init_leaderboard(dataframe):
             ColumnFilter(
                 AutoEvalColumn.params.name,
                 type="slider",
-                min=0.01,
+                min=0.00,
                 max=150,
                 label="Select the number of parameters (B)",
             ),
@@ -95,7 +95,7 @@ with demo:
     gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
 
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
-        with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0):
+        with gr.TabItem("🏅 Pinocchio ITA Benchmark", elem_id="llm-benchmark-tab-table", id=0):
             leaderboard = init_leaderboard(LEADERBOARD_DF)
 
         with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):

src/about.py

@@ -27,11 +27,22 @@ NUM_FEWSHOT = 0 # Change with your few shot
 
 
 # Your leaderboard name
-TITLE = """<h1 align="center" id="space-title">Demo leaderboard</h1>"""
+TITLE = """<h1 align="center" id="space-title">🇮🇹 Pinocchio ITA leaderboard from <a href="https://mii-llm.ai">mii-llm</a>🇮🇹</h1>"""
 
 # What does your leaderboard evaluate?
 INTRODUCTION_TEXT = """
-Intro 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. 
+The open source models are evaluated on the following subjects on Pinocchio tasks: 
+<ul>
+    <li>Generale</li>
+    <li>Logica</li>
+    <li>Lingua straniera</li>
+    <li>Matematica e scienze</li>
+    <li>Diritto</li>
+    <li>Cultura</li>
+</ul> 
 """
 
 # Which evaluations are you running? how can people reproduce what you have?

src/display/utils.py

@@ -64,7 +64,8 @@ class ModelDetails:
 class ModelType(Enum):
     PT = ModelDetails(name="pretrained", symbol="🟢")
     FT = ModelDetails(name="fine-tuned", symbol="🔶")
-    IFT = ModelDetails(name="instruction-tuned", symbol="⭕")
+    #IFT = ModelDetails(name="instruction-tuned", symbol="⭕")
+    IFT = ModelDetails(name="merged", symbol="⭕")
     RL = ModelDetails(name="RL-tuned", symbol="🟦")
     Unknown = ModelDetails(name="", symbol="?")
 
@@ -79,8 +80,10 @@ class ModelType(Enum):
             return ModelType.PT
         if "RL-tuned" in type or "🟦" in type:
             return ModelType.RL
-        if "instruction-tuned" in type or "⭕" in type:
+        if "merged" in type or "⭕" in type:
             return ModelType.IFT
+        #if "instruction-tuned" in type or "⭕" in type:
+        #    return ModelType.IFT
         return ModelType.Unknown
 
 class WeightType(Enum):