Update README.md

README.md

@@ -31,15 +31,6 @@ This model is fine-tuned to engage in productive disagreement, overthinking, and
 - **License**: Apache 2.0
 - **Finetuning Approach**: Instruction tuning with friction-based reasoning examples
 
-### Training Data
-
-The model was trained on a combination of three datasets:
-1. `leonvanbokhorst/friction-disagreement-v2` (8.5% weight)
-   - Examples of productive disagreement and challenging assumptions
-2. `leonvanbokhorst/friction-overthinking-v2` (9.5% weight)
-   - Examples of deep analytical thinking and self-reflection
-3. `leonvanbokhorst/reluctance-v6.1` (82% weight)
-   - Examples of hesitation and careful consideration
 
 ### Training Procedure
 
@@ -54,12 +45,6 @@ The model was trained on a combination of three datasets:
 - **Gradient Clipping**: 0.5
 - **Mixed Precision**: bfloat16
 
-### Performance Metrics
-
-- **Training Loss**: 1.437 (final)
-- **Best Validation Loss**: 1.527 (epoch 3.57)
-- **Memory Usage**: 3.813 GB for training (15.9% of GPU memory)
-
 ## Intended Use
 
 This model is designed for: