update usage code

README.md

@@ -39,27 +39,18 @@ We would like to thank the Baseten team for their contributions in deploying and
 
 ## Usage :
 ```python
-import os 
 import torch
-
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
-# set HF_TOKEN in terminal as export HF_TOKEN=hf_***
-auth_token = os.environ.get("HF_TOKEN", True)
-
 model_name = "Writer/camel-5b-hf"
 
-tokenizer = AutoTokenizer.from_pretrained(
-   model_name, use_auth_token=auth_token
-)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(
-   model_name,
+    model_name,
     device_map="auto",
-    torch_dtype=torch.float16,
-    use_auth_token=auth_token,
+    torch_dtype=torch.float16
 )
 
-
 instruction = "Describe a futuristic device that revolutionizes space travel."
 
 
@@ -85,16 +76,14 @@ text = (
 model_inputs = tokenizer(text, return_tensors="pt").to("cuda")
 output_ids = model.generate(
     **model_inputs,
-    max_length=100,
+    max_length=256,
 )
 output_text = tokenizer.batch_decode(output_ids, skip_special_tokens=True)[0]
 clean_output = output_text.split("### Response:")[1].strip()
 
 print(clean_output)
-
 ```
 
-`
 
 ### Limitations and Biases