Example

from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "jaeyong2/QuerySense-Preview"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = """
# Role
You are an AI that receives Questions and Context from users as input and preprocesses the Questions.

# Instruction
- If the user's Questions contains enough information to create an answer, use the user's Questions as is.
- If the information is insufficient or the Context is insufficient, please rephrase the Questions with the necessary information.
- If there is insufficient information to generate an answer and there is no Context, it will automatically fill in the appropriate information.

# input
- Context : Previous conversations or related Context or related information entered by the user (Optional)
- Question : User's Questions (Required)
""".strip()

content ="""
Context :
Question : name
""".strip()

system = {"role":"system", "content":prompt}
user = {"role":"user", "content":content}
messages = [system, user]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=False # Switches between thinking and non-thinking modes. Default is True.
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# conduct text completion
generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
content = tokenizer.decode(output_ids, skip_special_tokens=True).strip("\n")

print("content:", content)

result

content: what is the name of the product?

License

Acknowledgement

This research is supported by TPU Research Cloud program.

Downloads last month
20
Safetensors
Model size
1.72B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support