improvements

ai_eval.py

@@ -929,15 +929,21 @@ Do not include any text before or after the JSON object."""
     
     def save_results(self):
         """Save results to JSON file"""
+        # Ensure output directory exists
+        self.output_dir.mkdir(parents=True, exist_ok=True)
+        
+        # Sanitize model name for use in filename (replace problematic characters)
+        safe_model_name = self.model_name.replace('/', '_').replace(':', '_')
+        
         timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-        filename = f"{self.model_name.replace(':', '_')}_{timestamp}.json"
+        filename = f"{safe_model_name}_{timestamp}.json"
         filepath = self.output_dir / filename
         
         with open(filepath, 'w', encoding='utf-8') as f:
             json.dump(self.results, f, indent=2, ensure_ascii=False)
         
         # Also save as "latest" for this model
-        latest_file = self.output_dir / f"{self.model_name.replace(':', '_')}_latest.json"
+        latest_file = self.output_dir / f"{safe_model_name}_latest.json"
         with open(latest_file, 'w', encoding='utf-8') as f:
             json.dump(self.results, f, indent=2, ensure_ascii=False)
     

analyze_results.py

@@ -507,18 +507,14 @@ class WebInterface:
                     dimensions = {
                         'logical_reasoning': ['Logic & Reasoning'],
                         'mathematical_ability': ['Mathematics & Calculation'],
-                        'instruction_following': ['Instruction Following', 'Multi-turn: Instruction Following'],
+                        'instruction_following': ['Instruction Following with Constraints'],
-                        'creativity': ['Creative Writing'],
+                        'creativity': ['Creative Writing with Constraints'],
                         'technical_knowledge': [
-                            'Code Generation', 
+                            'Code Generation - Advanced',
-                            'IT Forensics - File Systems',
+                            'Digital Forensics & Binary Analysis'
-                            'IT Forensics - Registry & Artifacts',
-                            'IT Forensics - Memory & Network',
-                            'IT Forensics - Timeline & Log Analysis'
                         ],
-                        'linguistic_nuance': ['Language Nuance', 'Multilingual Competence'],
+                        'linguistic_nuance': ['Language Understanding & Pragmatics'],
-                        'problem_solving': ['Problem Solving & Logistics'],
+                        'conversational_depth': ['Multi-turn Context & Instruction Handling']
-                        'conversational_depth': ['Multi-turn: Context Retention']
                     }
                     
                     model_metrics = {}