KLGR123/WEPO-gemma-2b

Below is the reference code for inference. First load the tokenizer and the model.

from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("KLGR123/WEPO-gemma-2b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("KLGR123/WEPO-gemma-2b", trust_remote_code=True).to('cuda:0')

Run a test-demo with random input.

messages = [
    {"role": "system", "content": "You are a web navigation intelligence who interacts with webpage environments to achieve human user intent."},
    {"role": "user", "content": "Who are you?"},
]

input_ids = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

terminators = [
    tokenizer.eos_token_id,
    tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = model.generate(
    input_ids,
    max_new_tokens=128,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.2,
    top_p=0.9,
)

response = outputs[0][input_ids.shape[-1]:]
output = tokenizer.decode(response, skip_special_tokens=True)
output