JustinLin610
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -46,7 +46,10 @@ Here provides a code snippet with `apply_chat_template` to show you how to load
|
|
46 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
47 |
device = "cuda" # the device to load the model onto
|
48 |
|
49 |
-
model = AutoModelForCausalLM.from_pretrained(
|
|
|
|
|
|
|
50 |
tokenizer = AutoTokenizer.from_pretrained("Qwen2/Qwen2-beta-4B-Chat")
|
51 |
|
52 |
prompt = "Give me a short introduction to large language model."
|
@@ -55,7 +58,8 @@ messages = [
|
|
55 |
{"role": "user", "content": prompt}
|
56 |
]
|
57 |
text = tokenizer.apply_chat_template(
|
58 |
-
messages,
|
|
|
59 |
add_generation_prompt=True
|
60 |
)
|
61 |
model_inputs = tokenizer([text], return_tensors="pt").to(device)
|
|
|
46 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
47 |
device = "cuda" # the device to load the model onto
|
48 |
|
49 |
+
model = AutoModelForCausalLM.from_pretrained(
|
50 |
+
"Qwen2/Qwen2-beta-4B-Chat",
|
51 |
+
device_map="auto"
|
52 |
+
)
|
53 |
tokenizer = AutoTokenizer.from_pretrained("Qwen2/Qwen2-beta-4B-Chat")
|
54 |
|
55 |
prompt = "Give me a short introduction to large language model."
|
|
|
58 |
{"role": "user", "content": prompt}
|
59 |
]
|
60 |
text = tokenizer.apply_chat_template(
|
61 |
+
messages,
|
62 |
+
tokenize=False,
|
63 |
add_generation_prompt=True
|
64 |
)
|
65 |
model_inputs = tokenizer([text], return_tensors="pt").to(device)
|