Spaces:
QLWD
/
Running on Zero

vilarin commited on
Commit
639e063
1 Parent(s): 997f90e

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +2 -1
app.py CHANGED
@@ -64,7 +64,8 @@ def stream_chat(message: str, history: list, temperature: float, max_new_tokens:
64
  conversation.append({"role": "user", "content": message})
65
 
66
  input_ids = tokenizer.apply_chat_template(conversation, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
67
-
 
68
  gen_tokens= model.generate(
69
  input_ids,
70
  max_new_tokens=max_new_tokens,
 
64
  conversation.append({"role": "user", "content": message})
65
 
66
  input_ids = tokenizer.apply_chat_template(conversation, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
67
+ prompt_padded_len = len(input_ids[0])
68
+
69
  gen_tokens= model.generate(
70
  input_ids,
71
  max_new_tokens=max_new_tokens,