Next_token ambiguity in Causal Language Modeling sample

I am using the sample following sample code for Causal LM:

from transformers import AutoModelWithLMHead, AutoTokenizer, top_k_top_p_filtering
import torch
from torch.nn import functional as F

tokenizer = AutoTokenizer.from_pretrained("gpt2")
model = AutoModelWithLMHead.from_pretrained("gpt2")

sequence = f"Hugging Face is based in DUMBO, New York City, and "

input_ids = tokenizer.encode(sequence, return_tensors="pt")

# get logits of last hidden state
next_token_logits = model(input_ids)[0][:, -1, :]

# filter
filtered_next_token_logits = top_k_top_p_filtering(next_token_logits, top_k=50, top_p=1.0)

# sample
probs = F.softmax(filtered_next_token_logits, dim=-1)
next_token = torch.multinomial(probs, num_samples=1)

generated =[input_ids, next_token], dim=-1)

resulting_string = tokenizer.decode(generated.tolist()[0])

The next token generated by the sample code is has whereas when I run the same code on my local system, the next token generated in \xa0. Seems like somewhere version mismatch about the vocabulary be there. Can anyone give a precise answer about this ambiguity?

Update: transformers version on my local system: 4.6.1