update readme

README.md

@@ -47,6 +47,8 @@ for i in range(len(outputs)):
     print(f'{i + 1}: {query}')
 ```
 
+**Note:** `model.generate()` is non-deterministic. It produces different queries each time you run it.
+
 ## Training
 This model fine-tuned [google/t5-v1_1-base](https://huggingface.co/google/t5-v1_1-base) for 575k training steps. For the  training script, see the `train_script.py` in this repository.
 
@@ -64,7 +66,7 @@ The datasets include besides others:
 
 ## Prefix
 
-This model was trained **with prefixed**: You start the text with a specific index that defines what type out output text you would like to receive. Depending on the prefix, the output is different.
+This model was trained **with a prefix**: You start the text with a specific index that defines what type out output text you would like to receive. Depending on the prefix, the output is different.
 
 E.g. the above text about Python produces the following output: