Update README.md

README.md

@@ -57,7 +57,7 @@ The format for TinyCot was:
 
 For the initial supervised finetuning step:
 - Adalite optimizer, default hyperparameters of supertrainer2000 unless otherwise specified
-- Lambda (Adalite's analogue to weight decay) of 0.01
+- Lambda (Adalite's analogue to weight decay, see [here](https://arxiv.org/abs/2103.06583) for details) of 0.01
 - LR of 1e-5
 - MixCE ratio of 0.75
 - Sequence length of 4096