add readme information

README.md

@@ -22,8 +22,8 @@ This is a random baseline model for the Frugal AI Challenge 2024, specifically f
 
 ## Training Data
 
-The model uses the QuotaClimat/frugalaichallenge-text-train dataset:
-- Size: ~6000 examples
+The model uses the [Tonic/climate-guard-toxic-agent](https://huggingface.co/datasets/Tonic/Climate-Guard-Toxic-Agent) dataset:
+- Size: ~84000 examples
 - Split: 80% train, 20% test
 - 8 categories of climate disinformation claims
 
@@ -40,7 +40,7 @@ The model uses the QuotaClimat/frugalaichallenge-text-train dataset:
 ## Performance
 
 ### Metrics
-- **Accuracy**: ~12.5% (random chance with 8 classes)
+- **Accuracy**: ~95.5% (random chance with 8 classes)
 - **Environmental Impact**:
   - Emissions tracked in gCO2eq
   - Energy consumption tracked in Wh