rphrp1985 commited on
Commit
e4daf96
1 Parent(s): b4080b6

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +2 -2
app.py CHANGED
@@ -80,8 +80,8 @@ model = AutoModelForCausalLM.from_pretrained(model_id, token= token,
80
  # torch_dtype=torch.float16,
81
  load_in_8bit=True,
82
  # torch_dtype=torch.fl,
83
- # attn_implementation="flash_attention_2",
84
- # low_cpu_mem_usage=True,
85
  # device_map='cuda',
86
  # device_map=accelerator.device_map,
87
 
 
80
  # torch_dtype=torch.float16,
81
  load_in_8bit=True,
82
  # torch_dtype=torch.fl,
83
+ attn_implementation="flash_attention_2",
84
+ low_cpu_mem_usage=True,
85
  # device_map='cuda',
86
  # device_map=accelerator.device_map,
87