Add loss graph

README.md

@@ -29,16 +29,18 @@ pipeline_tag: text-generation
 This model is inspired by OpenAI's o1 reasoning model. The dataset was synthetically generated using Claude 3.5 Sonnet.
 
 ### Training
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/6604e5b21eb292d6df393365/pGUUKI77dP9NBVdLf1zB0.png)
 
-The training was done on Google Colab's free T4, using unsloth. The configuration is as follows:
+The training was done on Google Colab's free T4, using unsloth (duration: 52.32 minutes). The configuration is as follows:
 - LoRA Rank: 128
 - Packing: enabled
 - Batch size: 2
 - Gradient accumulation steps: 4
 - Epoches: 3
-- Steps: 240
+- Steps: 30
+- Max sequence length: 4096
 
-The training data comprised of 81 examples, each approximatly 3000 tokens. 
+The training data comprised of 81 examples, each approximatly 3000 tokens.
 
 ### Notes
 - It tends to produce very verbose and long reasoning responses