(no commit message)

config.json

@@ -1,6 +1,6 @@
 {
   "model": "openai//models/checkpoint-5",
   "api_base": "https://modaic-ai--grpo-demo-serve.modal.run/v1",
-  "max_tokens": 1024,
-  "temperature": 0.7
+  "max_tokens": 10000,
+  "temperature": 2
 }

main.py

@@ -4,8 +4,8 @@ from modaic import PrecompiledProgram, PrecompiledConfig
 class CodeGeneratorConfig(PrecompiledConfig):
     model : str = "openai//models/checkpoint-5"
     api_base : str = "https://modaic-ai--grpo-demo-serve.modal.run/v1"
-    max_tokens : int = 1024
-    temperature : float = 0.7
+    max_tokens : int = 10000
+    temperature : float = 2
 
 
 class CodeGeneration(dspy.Signature):
@@ -32,6 +32,6 @@ class CodeGenerator(PrecompiledProgram):
 
 code_generator = CodeGenerator(CodeGeneratorConfig())
 print(code_generator(query="Write a python function that returns the sum of two numbers.").code)
-code_generator.push_to_hub("modaic/code-generator-trl-grpo", with_code=True, tag="v1")
+code_generator.push_to_hub("modaic/code-generator-trl-grpo", with_code=True, tag="v2")
     
     

program.json

@@ -24,8 +24,8 @@
       "finetuning_model": null,
       "launch_kwargs": {},
       "train_kwargs": {},
-      "temperature": 0.7,
-      "max_tokens": 1024,
+      "temperature": 2,
+      "max_tokens": 10000,
       "api_base": "https://modaic-ai--grpo-demo-serve.modal.run/v1"
     }
   },