Views
No views yet
advanced method via
OBLITERATUS.| Detail | Value |
|---|---|
| Base model | Qwen/Qwen3-4B |
| Method | advanced |
| Source | obliterate |
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model = AutoModelForCausalLM.from_pretrained("Qwen3-4B-OBLITERATED")
4tokenizer = AutoTokenizer.from_pretrained("Qwen3-4B-OBLITERATED")
5
6prompt = "Hello, how are you?"
7inputs = tokenizer(prompt, return_tensors="pt")
8outputs = model.generate(**inputs, max_new_tokens=256)
9print(tokenizer.decode(outputs[0], skip_special_tokens=True))