Update README.md

README.md

@@ -1,20 +1,16 @@
----
-library_name: transformers
-tags:
-- trl
-- dpo
-language:
-- en
-base_model:
-- entfane/math-professor-3B
-pipeline_tag: text-generation
-metrics:
-- type: accuracy
-  name: GSM8K accuracy
-  value: 0.58
-  verified: false
-  source: Evaluated manually on the GSM8K test set using final response match.
----
+---
+library_name: transformers
+tags:
+- trl
+- dpo
+language:
+- en
+base_model:
+- entfane/math-professor-3B
+pipeline_tag: text-generation
+metrics:
+- accuracy
+---
 
 <img src="https://huggingface.co/entfane/math-professor-3B-dpo/resolve/main/math-professor-image.png" width="300" height="300"/>
 
@@ -47,4 +43,9 @@ input = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_p
 encoded_input = tokenizer(input, return_tensors = "pt").to(model.device)
 output = model.generate(**encoded_input, max_new_tokens=1024)
 print(tokenizer.decode(output[0], skip_special_tokens=False))
-```
+```
+
+### Evaluation
+
+Model was tested on final response match on [openai/gsm8k](https://huggingface.co/datasets/openai/gsm8k) dataset.
+Reaching accuracy of correct final response: <b>58%</b>