Update README.md

README.md

@@ -14,17 +14,12 @@ In addition to cross entropy and cosine teacher-student losses, DistilProtBert w
 
 
 Access to [git](https://github.com/yarongef/DistilProtBert)
-# Model description
+# DistilProtBert comparison to ProtBert
 
-DistilProtBert was pretrained on millions of proteins sequences. 
-
-
-Differences between DistilProtBert model and ProtBert:
-
-|    **Model**   | **# of Parameters** | **# of Hidden layers** | **# of Pretraining sequences** | **Pretraining hardware** |
-|:--------------:|:--------------:|:-----------------:|:-------------------------:|:------------------------:|
-|    ProtBert    |      420M      |         30        |            216M            |     512 16GB TPUs      |
-| DistilProtBert |      230M      |         15        |            43M           |       5 v100 32GB GPUs      |
+|    **Model**   | **# of parameters** | **# of hidden layers** | **Pretraining dataset** | **# of pretraining sequences** | **Pretraining hardware** |
+|:--------------:|:-------------------:|:----------------------:|:-----------------------:|:------------------------------:|:------------------------:|
+|    ProtBert    |         420M        |           30           |        UniRef100        |              216M              |       512 16GB Tpus      |
+| DistilProtBert |         230M        |           15           |         UniRef50        |               43M              |     5 v100 32GB GPUs     |
 
 ## Intended uses & limitations