Update README.md

README.md

@@ -27,15 +27,16 @@ If any of these two is not installed, the "eager" implementation will be used. O
 ## Generation
 You can use the classic `generate` API:
 ```python
-from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer
-import torch
+>>> from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer
+>>> import torch
 
-tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-2.8b-hf")
-model = MambaForCausalLM.from_pretrained("state-spaces/mamba-2.8b-hf")
-input_ids = tokenizer("Hey how are you doing?", return_tensors="pt")["input_ids"]
+>>> tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-2.8b-hf")
+>>> model = MambaForCausalLM.from_pretrained("state-spaces/mamba-2.8b-hf")
+>>> input_ids = tokenizer("Hey how are you doing?", return_tensors="pt")["input_ids"]
 
-out = model.generate(input_ids, max_new_tokens=10)
-print(tokenizer.batch_decode(out))
+>>> out = model.generate(input_ids, max_new_tokens=10)
+>>> print(tokenizer.batch_decode(out))
+["Hey how are you doing?\n\nI'm doing great.\n\nI"]
 ```
 
 ## PEFT finetuning example