Spaces:
Runtime error
Runtime error
FlawedLLM
commited on
Update app.py
Browse files
app.py
CHANGED
@@ -64,13 +64,13 @@ from huggingface_hub import login, HfFolder
|
|
64 |
# model = AutoModelForCausalLM.from_pretrained("FlawedLLM/Bhashini_9",config=config, ignore_mismatched_sizes=True).to('cuda')
|
65 |
# Load model directly
|
66 |
|
67 |
-
tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/
|
68 |
quantization_config = BitsAndBytesConfig(
|
69 |
load_in_4bit=True,
|
70 |
bnb_4bit_use_double_quant=True,
|
71 |
bnb_4bit_quant_type="nf4",
|
72 |
bnb_4bit_compute_dtype=torch.float16)
|
73 |
-
model = AutoModelForCausalLM.from_pretrained("FlawedLLM/
|
74 |
device_map="auto",
|
75 |
quantization_config=quantization_config,
|
76 |
torch_dtype =torch.float16,
|
|
|
64 |
# model = AutoModelForCausalLM.from_pretrained("FlawedLLM/Bhashini_9",config=config, ignore_mismatched_sizes=True).to('cuda')
|
65 |
# Load model directly
|
66 |
|
67 |
+
tokenizer = AutoTokenizer.from_pretrained("FlawedLLM/Bhashini89", trust_remote_code=True)
|
68 |
quantization_config = BitsAndBytesConfig(
|
69 |
load_in_4bit=True,
|
70 |
bnb_4bit_use_double_quant=True,
|
71 |
bnb_4bit_quant_type="nf4",
|
72 |
bnb_4bit_compute_dtype=torch.float16)
|
73 |
+
model = AutoModelForCausalLM.from_pretrained("FlawedLLM/Bhashini89",
|
74 |
device_map="auto",
|
75 |
quantization_config=quantization_config,
|
76 |
torch_dtype =torch.float16,
|