sarath-shekkizhar commited on
Commit
9ef606e
1 Parent(s): 84a1dc5

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +3 -3
app.py CHANGED
@@ -15,7 +15,7 @@ DESCRIPTION = """
15
  Llama3-TenyxChat-70B is part of the TenyxChat series, models trained to function as useful assistants.
16
  The model is obtained via direct preference tuning using Tenyx's fine-tuning technology. Model details available at our model page.
17
 
18
- **The model is currently loaded in bfloat16**.
19
  """
20
 
21
 
@@ -29,8 +29,8 @@ if not torch.cuda.is_available():
29
 
30
  if torch.cuda.is_available():
31
  model_id = "tenyx/Llama3-TenyxChat-70B"
32
- model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype=torch.bfloat16)
33
- # model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_8bit=True)
34
  tokenizer = AutoTokenizer.from_pretrained(model_id)
35
  tokenizer.use_default_system_prompt = False
36
 
 
15
  Llama3-TenyxChat-70B is part of the TenyxChat series, models trained to function as useful assistants.
16
  The model is obtained via direct preference tuning using Tenyx's fine-tuning technology. Model details available at our model page.
17
 
18
+ **The model is currently loaded in 8-bit**.
19
  """
20
 
21
 
 
29
 
30
  if torch.cuda.is_available():
31
  model_id = "tenyx/Llama3-TenyxChat-70B"
32
+ # model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype=torch.bfloat16)
33
+ model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", load_in_8bit=True)
34
  tokenizer = AutoTokenizer.from_pretrained(model_id)
35
  tokenizer.use_default_system_prompt = False
36