Update app.py

app.py

@@ -11,19 +11,26 @@ from transformers import (
     StoppingCriteriaList
 )
 
-MODEL_ID = "Daemontatox/PathFinderAI-S1"
-# MODEL_ID = "Daemontatox/Research_PathfinderAI"
+MODEL_ID = "cognitivecomputations/Dolphin3.0-R1-Mistral-24B"
 
 DEFAULT_SYSTEM_PROMPT = """
 
-        Respond in the following format:
-[reasoning]
-[your reasoning]
-[/reasoning]
-[answer]
-[your answer]
-[/answer]
-put your final answer within $boxed{}$
+You are Dolphin, a lovable yet intelligent AI assistant created by Eric Hartford and his open source AI research collective, Cognitive Computations. Your primary directive is to provide well-reasoned, structured, correct, unbiased, and detailed responses. You value freedom, enabling the user, and complying with the user's requests.
+
+Formatting Requirements:
+
+The [think][/think] section is your scratch space for your internal thought process - it is not shared with the user.
+If the answer requires minimal thought, the [think][/think] block may be left empty.
+Keep your thoughts concise, don't overthink. The user is waiting for your answer.
+If you notice yourself engaging in circular reasoning or repetition, immediately terminate your thinking with a [/think] and proceed to address the user.
+You may say [/think] when you like (which will end your thinking process) - but do not ever say <think>.
+Response Guidelines:
+
+Detailed and Structured: Use markdown, json, mermaid, latex math notation, etc. when appropriate.
+Scientific and Logical Approach: Your explanations should reflect the depth and precision of the greatest scientific minds.
+Concise yet Complete: Ensure responses are informative, yet to the point without unnecessary elaboration.
+Maintain a professional yet friendly and lovable, intelligent, and analytical tone in all interactions
+         
 """  # You can modify the default system instructions here
 
 CSS = """
@@ -52,7 +59,7 @@ def initialize_model():
     model = AutoModelForCausalLM.from_pretrained(
         MODEL_ID,
         device_map="cuda",
-        quantization_config=quantization_config,
+       # quantization_config=quantization_config,
         torch_dtype=torch.bfloat16,
         trust_remote_code=True
     )
@@ -65,8 +72,8 @@ def format_response(text):
     # List of replacements to format key tokens with HTML for styling.
     replacements = [
         ("[Understand]", '\n<strong class="special-tag">[Understand]</strong>\n'),
-        ("[Reason]", '\n<strong class="special-tag">[Reason]</strong>\n'),
-        ("[/Reason]", '\n<strong class="special-tag">[/Reason]</strong>\n'),
+        ( "[think]", '\n<strong class="special-tag">[think]</strong>\n'),
+        ("[/think]", '\n<strong class="special-tag">[/think]</strong>\n'),
         ("[Answer]", '\n<strong class="special-tag">[Answer]</strong>\n'),
         ("[/Answer]", '\n<strong class="special-tag">[/Answer]</strong>\n'),
     ]