Update code examples so they can work out of the box

README.md

@@ -35,6 +35,7 @@ This allows the model to better adapt to human preferences in the conversations.
 This requires a GPU with 48GB memory.
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 # init
 tokenizer = AutoTokenizer.from_pretrained("togethercomputer/GPT-NeoXT-Chat-Base-20B")
 model = AutoModelForCausalLM.from_pretrained("togethercomputer/GPT-NeoXT-Chat-Base-20B", torch_dtype=torch.float16)
@@ -66,6 +67,7 @@ print(output_str)
 
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 # init
 tokenizer = AutoTokenizer.from_pretrained("togethercomputer/GPT-NeoXT-Chat-Base-20B")
 model = AutoModelForCausalLM.from_pretrained("togethercomputer/GPT-NeoXT-Chat-Base-20B", torch_dtype=torch.bfloat16)