Usage
!pip install -q -U trl transformers accelerate git+https://github.com/huggingface/peft.git
!pip install -q datasets bitsandbytes einops wandb sentencepiece transformers_stream_generator tiktoken
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
tokenizer = AutoTokenizer.from_pretrained("TinyPixel/qwen-1.8B-OrcaMini", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("TinyPixel/qwen-1.8B-OrcaMini", torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)
device = "cuda:0"
text = '''SYSTEM:
USER: what is the difference between a dog and a cat on a biological level?
ASSISTANT:'''
inputs = tokenizer(text, return_tensors="pt").to(device)
outputs = model.generate(**inputs,
max_new_tokens=512,
do_sample=True,
top_p=0.95,
temperature=0.7,
top_k=50)
print(tokenizer.decode(outputs[0], skip_special_tokens=False))
