Update README.md

README.md

@@ -33,6 +33,7 @@ TODO: get model architecture like on the model card of the previous model
 
 Full architecture:
 
+  ```
   [
     {
       "idx": 0,
@@ -47,9 +48,11 @@ Full architecture:
       "type": "sentence_transformers.sentence_transformer.modules.pooling.Pooling"
     }
   ]
+```
 
 Parameters used: 
 
+```
   {
     "architectures": [
       "BertForMaskedLM"
@@ -74,9 +77,11 @@ Parameters used:
     "use_cache": true,
     "vocab_size": 6161
   }
+```
 
 Preraining parameters used: 
 
+```
   {
     "epochs"=19, 
     "eval_steps"=10000, 
@@ -86,6 +91,7 @@ Preraining parameters used:
     "bf16"=True, 
     "tf32"=True
   }
+```
 
 Note that during pretraining, the --mlm-prob parameter has been set to 0.4, in accordance to [paper](INSERT LINK)