Aquila2-34B / README.md
shunxing1234's picture
Update README.md
3917ce7
|
raw
history blame
No virus
2.79 kB
metadata
license: other

Aquila_logo

English | 简体中文

We opensource our Aquila2 series, now including Aquila2, the base language models, namely Aquila2-7B and Aquila2-34B, as well as AquilaChat2, the chat models, namely AquilaChat2-7B and AquilaChat2-34B, as well as the long-text chat models, namely AquilaChat2-7B-16k and AquilaChat2-34B-16k

The additional details of the Aquila model will be presented in the official technical report. Please stay tuned for updates on official channels.

2023.10.25 🔥 Version 1.2 of Aquila2-34B, AquilaChat2-34B-16K and AquilaChat2-34B models have been updated on ModelHub and Hugging Face. Among them, the Aquila2-34B model has improved by 6.9% in comprehensive objective evaluation. The evaluation results of Aquila2-34B v1.2 on MMLU, TruthfulQA, CSL, TNEWS, OCNLI, BUSTM and other exam, understanding and reasoning datasets have respectively increased by 12%, 14%, 11%, 12%, 28%, 18%.

Chat Model Performance



Quick Start Aquila2-34B(Chat model)

1. Inference

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
from transformers import BitsAndBytesConfig

device = torch.device("cuda")
model_info = "BAAI/Aquila2-34B"
tokenizer = AutoTokenizer.from_pretrained(model_info, trust_remote_code=True)
quantization_config=BitsAndBytesConfig(
                        load_in_4bit=True,
                        bnb_4bit_use_double_quant=True,
                        bnb_4bit_quant_type="nf4",
                        bnb_4bit_compute_dtype=torch.bfloat16,
                    )
model = AutoModelForCausalLM.from_pretrained(model_info, trust_remote_code=True, 
                                                # quantization_config=quantization_config, # Uncomment this line for 4bit quantization
                                                )
model.eval()
model.to(device)
text = "请给出10个要到北京旅游的理由。"
tokens = tokenizer.encode_plus(text)['input_ids']
tokens = torch.tensor(tokens)[None,].to(device)
stop_tokens = ["###", "[UNK]", "</s>"]
with torch.no_grad():
    out = model.generate(tokens, do_sample=True, max_length=512, eos_token_id=100007, bad_words_ids=[[tokenizer.encode(token)[0] for token in stop_tokens]])[0]
    out = tokenizer.decode(out.cpu().numpy().tolist())
    print(out)

License

Aquila2 series open-source model is licensed under BAAI Aquila Model Licence Agreement