import torch from transformers import AutoTokenizer, AutoModelForSequenceClassification model_name = "vraj33/ai-text-detector-deberta" tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForSequenceClassification.from_pretrained(model_name) print(f"Model: {model_name}") print(f"Config id2label: {model.config.id2label}") texts = ["This is a human written sentence.", "The artificial intelligence system generated this response."] for text in texts: inputs = tokenizer(text, return_tensors="pt") outputs = model(**inputs) probs = torch.softmax(outputs.logits, dim=-1) print(f"Text: {text}") print(f"Probs: {probs}")