Edit model card

MiniChat-3B

πŸ“‘ arXiv | πŸ‘» GitHub | πŸ€— HuggingFace-MiniMA | πŸ€— HuggingFace-MiniChat | πŸ€— HuggingFace-MiniChat-1.5 | πŸ€– ModelScope-MiniMA | πŸ€– ModelScope-MiniChat

πŸ†• Updates: MiniChat-1.5-3B

❗ Must comply with LICENSE of LLaMA2 since it is derived from LLaMA2.

A language model distilled and finetuned from an adapted version of LLaMA2-7B following "Towards the Law of Capacity Gap in Distilling Language Models".

Outperforming a wide range of 3B competitors in GPT4 evaluation and even competing with several 7B chat models.

teaser_b

The following is an example code snippet to use MiniChat-3B:

import torch

from transformers import AutoModelForCausalLM, AutoTokenizer

from conversation import get_default_conv_template

# MiniChat
tokenizer = AutoTokenizer.from_pretrained("GeneZC/MiniChat-3B", use_fast=False)
# GPU.
model = AutoModelForCausalLM.from_pretrained("GeneZC/MiniChat-3B", use_cache=True, device_map="auto", torch_dtype=torch.float16).eval()
# CPU.
# model = AutoModelForCausalLM.from_pretrained("GeneZC/MiniChat-3B", use_cache=True, device_map="cpu", torch_dtype=torch.float32).eval()
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
conv = get_default_conv_template("minichat")

question = "Implement a program to find the common elements in two arrays without using any extra data structures."
conv.append_message(conv.roles[0], question)
conv.append_message(conv.roles[1], None)
prompt = conv.get_prompt()
input_ids = tokenizer([prompt]).input_ids
output_ids = model.generate(
    torch.as_tensor(input_ids).to(device),
    do_sample=True,
    temperature=0.7,
    max_new_tokens=1024,
)
output_ids = output_ids[0][len(input_ids[0]):]
output = tokenizer.decode(output_ids, skip_special_tokens=True).strip()
# output: "def common_elements(arr1, arr2):\n    if len(arr1) == 0:\n        return []\n    if len(arr2) == 0:\n        return arr1\n\n    common_elements = []\n    for element in arr1:\n        if element in arr2:\n            common_elements.append(element)\n\n    return common_elements"
# Multiturn conversation could be realized by continuously appending questions to `conv`.

Bibtex

@article{zhang2023law,
    title={Towards the Law of Capacity Gap in Distilling Language Models},
    author={Zhang, Chen and Song, Dawei and Ye, Zheyu and Gao, Yan},
    year={2023},
    url={https://arxiv.org/abs/2311.07052}
}

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 42.94
ARC (25-shot) 44.03
HellaSwag (10-shot) 67.19
MMLU (5-shot) 39.17
TruthfulQA (0-shot) 45.67
Winogrande (5-shot) 65.27
GSM8K (5-shot) 10.54
DROP (3-shot) 28.73
Downloads last month
3,977
Safetensors
Model size
3.02B params
Tensor type
BF16
Β·

Spaces using GeneZC/MiniChat-3B 4