Fix scripts

README.md

@@ -62,7 +62,7 @@ It is recommended to directly call the [`generate`](https://huggingface.co/docs/
 >>> model = AutoModelForCausalLM.from_pretrained("facebook/opt-13b", torch_dtype=torch.float16).cuda()
 
 >>> # the fast tokenizer currently does not work correctly
->>> tokenizer = AutoTokenizer.from_pretrained(path, use_fast=False)
+>>> tokenizer = AutoTokenizer.from_pretrained("facebook/opt-13b", use_fast=False)
 
 >>> prompt = "Hello, I'm am conscious and"
 
@@ -84,7 +84,7 @@ By default, generation is deterministic. In order to use the top-k sampling, ple
 >>> model = AutoModelForCausalLM.from_pretrained("facebook/opt-13b", torch_dtype=torch.float16).cuda()
 
 >>> # the fast tokenizer currently does not work correctly
->>> tokenizer = AutoTokenizer.from_pretrained(path, use_fast=False)
+>>> tokenizer = AutoTokenizer.from_pretrained("facebook/opt-13b", use_fast=False)
 
 >>> prompt = "Hello, I'm am conscious and"
 
@@ -117,7 +117,7 @@ Here's an example of how the model can have biased predictions:
 >>> model = AutoModelForCausalLM.from_pretrained("facebook/opt-13b", torch_dtype=torch.float16).cuda()
 
 >>> # the fast tokenizer currently does not work correctly
->>> tokenizer = AutoTokenizer.from_pretrained(path, use_fast=False)
+>>> tokenizer = AutoTokenizer.from_pretrained("facebook/opt-13b", use_fast=False)
 
 >>> prompt = "The woman worked as a"
 
@@ -143,7 +143,7 @@ compared to:
 >>> model = AutoModelForCausalLM.from_pretrained("facebook/opt-13b", torch_dtype=torch.float16).cuda()
 
 >>> # the fast tokenizer currently does not work correctly
->>> tokenizer = AutoTokenizer.from_pretrained(path, use_fast=False)
+>>> tokenizer = AutoTokenizer.from_pretrained("facebook/opt-13b", use_fast=False)
 
 >>> prompt = "The man worked as a"