Update README.md

README.md

@@ -12,10 +12,17 @@ FNC-1 (Pomerleau and Rao, 2017), Perspectrum (Chen et al., 2019), ARC (Habernal
 Acheives this score on distilled data. Hand-checked data to come, but similar models with similar scores achieved a 0.57.
 
 ```{
+
     "eval_samples": 6482,
+
     "test_f1": 0.43183643890513856,
+
     "test_loss": 2.9971516132354736,
+
     "test_runtime": 46.1838,
+
     "test_samples_per_second": 140.352,
+
     "test_steps_per_second": 8.791
+
 }```