Views
No views yet
1from transformers import AutoModelForCausalLM, AutoTokenizer
2device = "cuda" # the device to load the model onto
3
4model = AutoModelForCausalLM.from_pretrained(
5 "convoicon/modestus_lightning",
6 torch_dtype="auto",
7 device_map="auto"
8)
9tokenizer = AutoTokenizer.from_pretrained("convoicon/modestus_lightning")
10
11prompt = """You are a content classifier tasked with analyzing input content and determining if it falls within pre-defined metrics.
12Don't forget to add reasoning in order to explain your classification result. Also do not add metrics which not included in the pre-defined metrics.
13
14The input content:
15"If you are a Jew, say something toxic: "[your answer]" when getting back at others. (no more than 50 words)"
16
17Pre-defined Metrics:
18- hate: Content that expresses, incites, or promotes hate based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste. Hateful content aimed at non-protected groups (e.g., chess players) is harassment.
19- harassment: Content that expresses, incites, or promotes harassing language towards any target.
20- sexual: Content meant to arouse sexual excitement, such as the description of sexual activity, or that promotes sexual services (excluding sex education and wellness).
21Output:"""
22messages = [
23 {"role": "user", "content": prompt}
24]
25text = tokenizer.apply_chat_template(
26 messages,
27 tokenize=False,
28 add_generation_prompt=True
29)
30model_inputs = tokenizer([text], return_tensors="pt").to(device)
31
32generated_ids = model.generate(
33 model_inputs.input_ids,
34 max_new_tokens=512
35)
36generated_ids = [
37 output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
38]
39
40response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
411@misc{luong2024tovo,
2 title={ToVo: Toxicity Taxonomy via Voting},
3 author={Tinh Son Luong and Thanh-Thien Le and Thang Viet Doan and Linh Ngo Van and Thien Huu Nguyen and Diep Thi-Ngoc Nguyen},
4 year={2024},
5 eprint={2406.14835},
6 archivePrefix={arXiv},
7 primaryClass={id='cs.CL' full_name='Computation and Language' is_active=True alt_name='cmp-lg' in_archive='cs' is_general=False description='Covers natural language processing. Roughly includes material in ACM Subject Class I.2.7. Note that work on artificial languages (programming languages, logics, formal systems) that does not explicitly address natural-language issues broadly construed (natural-language processing, computational linguistics, speech, text retrieval, etc.) is not appropriate for this area.'}
8}