File size: 4,445 Bytes
d6fa761
 
6f642b0
 
 
 
 
 
d6fa761
6f642b0
6cb84b1
6f642b0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2061213
 
 
 
 
 
 
 
 
 
 
6f642b0
2061213
 
 
 
 
 
6f642b0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
---
license: apache-2.0
language:
- en
- zh
library_name: transformers
widget:
- text: "<s> [|User|] Hi 👋  </s>[|Assistant|]"
---

## MiniChat-2-3B

📑 [arXiv](https://arxiv.org/abs/2311.07052) | 👻 [GitHub](https://github.com/GeneZC/MiniMA) | 🤗 [HuggingFace-MiniMA](https://huggingface.co/GeneZC/MiniMA-3B) | 🤗 [HuggingFace-MiniChat](https://huggingface.co/GeneZC/MiniChat-3B) | 🤖 [ModelScope-MiniMA](https://modelscope.cn/models/GeneZC/MiniMA-3B) | 🤖 [ModelScope-MiniChat](https://modelscope.cn/models/GeneZC/MiniChat-3B) | 🤗 [HuggingFace-MiniChat-1.5](https://huggingface.co/GeneZC/MiniChat-1.5-3B) | 🤗 [HuggingFace-MiniMA-2](https://huggingface.co/GeneZC/MiniMA-2-3B) | 🤗 [HuggingFace-MiniChat-2](https://huggingface.co/GeneZC/MiniChat-2-3B)

🆕 **Updates from MiniChat-3B**: 
- better base model MiniMA-2-3B;
- better data mixture;
- use of [NEFTune](https://arxiv.org/abs/2310.05914);
- use of [DPO](https://arxiv.org/abs/2305.18290).

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

A language model continued from MiniMA-3B and finetuned on both instruction and preference data.

Surpassing Vicuna-7B and approximating LLaMA-2-Chat-7B on MT-Bench.

<img src="./teaser_b.jpg" alt="teaser_b" width="687" />

**Standard Benchmarks**

|Method|TFLOPs|MMLU (5-shot)|CEval (5-shot)|DROP (3-shot)|HumanEval (0-shot)|BBH (3-shot)|GSM8K (8-shot)|
|--|--|--|--|--|--|--|--|
|Mamba-2.8B|4.6E9|25.58|24.74|15.72|7.32|29.37|3.49|
|ShearedLLaMA-2.7B|0.8E9|26.97|22.88|19.98|4.88|30.48|3.56|
|BTLM-3B|11.3E9|27.20|26.00|17.84|10.98|30.87|4.55|
|StableLM-3B|72.0E9|44.75|31.05|22.35|15.85|32.59|10.99|
|Qwen-1.8B|23.8E9|44.05|54.75|12.97|14.02|30.80|22.97|
|Phi-2-2.8B|159.9E9|56.74|34.03|30.74|46.95|44.13|55.42|
|LLaMA-2-7B|84.0E9|46.00|34.40|31.57|12.80|32.02|14.10|
||
|MiniMA-3B|4.0E9|28.51|28.23|22.50|10.98|31.61|8.11|
|MiniChat-3B|4.0E9|38.40|36.48|22.58|18.29|31.36|29.72|
|MiniMA-2-3B|13.4E9|40.14|44.65|23.10|14.63|31.43|8.87|
|MiniChat-2-3B|13.4E9|46.17|43.91|30.26|22.56|34.95|38.13|

**Instruction-following Benchmarks**

|Method|AlpacaEval|MT-Bench|MT-Bench-ZH|
|--|--|--|--|
|GPT-4|95.28|9.18|8.96|
|Zephyr-7B-Beta|90.60|7.34|6.27<sup>#</sup>|
|Vicuna-7B|76.84|6.17|5.22<sup>#</sup>|
|LLaMA-2-Chat-7B|71.37|6.27|5.43<sup>#</sup>|
|Qwen-Chat-7B|-|-|6.24|
|Phi-2-DPO|81.37|-|1.59<sup>#</sup><sup>$</sup>|
|StableLM-Zephyr-3B|76.00|6.64|4.31<sup>#</sup>|
|Rocket-3B|79.75|6.56|4.07<sup>#</sup>|
|Qwen-Chat-1.8B|-|-|5.65|
||
|MiniChat-3B|48.82|-|-|
|MiniChat-2-3B|77.30|6.23|6.04|

<sup>#</sup> specialized mainly for English.

<sup>$</sup> finetuned without multi-turn instruction data.

The following is an example code snippet to use MiniChat-2-3B:

```python
import torch

from transformers import AutoModelForCausalLM, AutoTokenizer

from conversation import get_default_conv_template

# MiniChat
tokenizer = AutoTokenizer.from_pretrained("GeneZC/MiniChat-2-3B", use_fast=False)
# GPU.
model = AutoModelForCausalLM.from_pretrained("GeneZC/MiniChat-2-3B", use_cache=True, device_map="auto", torch_dtype=torch.float16).eval()
# CPU.
# model = AutoModelForCausalLM.from_pretrained("GeneZC/MiniChat-2-3B", use_cache=True, device_map="cpu", torch_dtype=torch.float16).eval()

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).cuda(),
    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

```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}
}
```