fix secret

Author: linhaowei1

examples/symbolic_regression/README.md

@@ -16,24 +16,10 @@ This example leverages **LLM-SRBench**, a benchmark specifically designed for La
 
 Follow these steps to set up and run the symbolic regression benchmark example:
 
-### 1. Configure API Secrets
+### 1. Configure API Keys
 
-You'll need to provide your API credentials for the language models used by OpenEvolve.
+THe API key is read from the environment `OPENAI_API_KEY` by default. The primary and secondary model we used in testing LLM-SRBench is `gpt-4o` and `o3`. You can check `create_config()` in `data_api.py`.
 
-- Create a `secrets.yaml` file in the example directory.
-- Add your API key and model preferences:
-
-YAML
-
-```
-# secrets.yaml
-api_key: <YOUR_OPENAI_API_KEY>
-api_base: "https://api.openai.com/v1"  # Or your custom endpoint
-primary_model: "gpt-4o"
-secondary_model: "o3" # Or another preferred model for specific tasks
-```
-
-Replace `<YOUR_OPENAI_API_KEY>` with your actual OpenAI API key.
 
 ### 2. Load Benchmark Tasks & Generate Initial Programs