Update README.md

README.md

@@ -11,11 +11,15 @@ reasoning tasks. The objective is to enable the model to interpret natural langu
 spatial relationships, directions, and locations and output actionable responses.
 The task addresses limitations in current LLMs, which often fail to perform precise spatial reasoning,
 such as determining relationships between points on a map, planning routes, or identifying locations
-based on bounding boxes. For this fine-tuning task the step-game dataset was used.
+based on bounding boxes. 
+
+# Training Data
+
+For this fine-tuning task the step-game dataset was used.
 This dataset follows the template of story, question and answer as depicted below.
 
+![Description of StepGame Training Data](https://huggingface.co/sareena/spatial_lora_mistral/blob/main/stepgame.jpg)
 
-# Training Data
 
 # Training Method