Update README.md

Update the license

README.md

@@ -1,5 +1,5 @@
 ---
-license: apache-2.0
+license: cc
 tags:
 - trl
 - transformers
@@ -9,8 +9,6 @@ datasets:
 - ProlificAI/social-reasoning-rlhf
 language:
 - en
-metrics:
-- accuracy
 pipeline_tag: conversational
 ---
 
@@ -83,7 +81,7 @@ outputs = model(**inputs, labels=inputs["input_ids"])
 
 If you want to use the model for inference in a gradio app, consider the following code:
 
-'''python
+```python
 from trl import AutoModelForSeq2SeqLMWithValueHead
 from transformers import pipeline, AutoTokenizer
 import torch
@@ -123,6 +121,6 @@ gr.ChatInterface(
     examples=examples
 ).launch()
 
-'''
+```
 
 Make sure to keep all the tensors on the same device (CPU/GPU).