Update README.md

README.md

@@ -69,10 +69,10 @@ Regression and classification tasks with ≤50 000 samples and ≤2000 features
 - Not tested for >50 000 samples or > 2000 features.
 
 ### Model Architecture
-Transformer with TabPFNv2-like alternating attention with 18-24 layers
+Transformer with TabPFNv2-like alternating attention with 24 layers.
 
 ### Training Data and Priors
-TabPFN-2.6 is trained purely on synthetic tabular tasks
+TabPFN-2.6 is trained purely on synthetic tabular tasks.
 
 ### Performance Benchmarks
 Evaluated on proprietary benchmark collection, TabArena, and RealCause (for a causal version), in each of which it yields new SOTA results.