--- pipeline_tag: text-generation license: apache-2.0 language: - zh --- # Model Card for Breeze-7B-Instruct-v0.1 Breeze-7B is a language model that builds upon the foundation of [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1), specifically enhanced for Traditional Chinese. [Breeze-7B-Base-v0.1](https://huggingface.co/MediaTek-Research/Breeze-7B-Base-v0.1) introduces an expanded vocabulary with additional 30,000 Traditional Chinese tokens and is pre-trained on a substantial dataset of 250GB of Traditional Chinese content. With the expanded vocabulary, the base model operates at twice the inference speed for Traditional Chinese characters compared to Mistral-7B. [See [Inference Performance](#inference-performance).] This achievement marks a significant milestone as it is the first instance of vocabulary expansion in a model tailored for Traditional Chinese. [Breeze-7B-Instruct-v0.1](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v0.1) derives from the base model Breeze-7B-Base-v0.1 and has undergone supervised fine-tuning with over 1 million instances to sharpen its capabilities. This fine-tuned model demonstrates impressive performance in benchmarks for both English and Traditional Chinese, surpassing the results of Taiwan-LLM-7B-v2.1-chat, Taiwan-LLM-13B-v2.0-chat and Qwen-7B-chat in Traditional Chinese assessments. It also excels in some benchmarks against Yi-6B-Chat. In English evaluations, Breeze-7B-Instruct-v0.1 shows comparable results to Mistral-7B-Instruct-v0.1 on the MMLU and MT-Bench benchmarks. [See [Chat Model Performance](#chat-model-performance).] [Breeze-7B-Instruct-64k-v0.1](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-64k-v0.1) is an extension to Breeze-7B-Instruct-v0.1 to enable 64k context length, which is equivalent to 88k Traditional Chinese characters. With minimal sacrifice in the performance of the regular benchmarks, Breeze-7B-Instruct-64k-v0.1 can solve tasks such as question answering and summarization on document-level inputs. [See [Long-context Performance](#long-context-performance).] *A project by the members (in alphabetical order): Chan-Jan Hsu 許湛然, Chang-Le Liu 劉昶樂, Feng-Ting Liao 廖峰挺, Po-Chun Hsu 許博竣, Yi-Chang Chen 陳宜昌, and the supervisor Da-Shan Shiu 許大山.* ## Features - Breeze-7B-Base-v0.1 - Expanding the vocabulary dictionary size from 32k to 62k to better support Traditional Chinese - 8k tokens context length - Breeze-7B-Instruct-v0.1 - Expanding the vocabulary dictionary size from 32k to 62k to better support Traditional Chinese - 8k tokens context length - Multi-turn dialogue (without special handling for harmfulness) - Breeze-7B-Instruct-64k-v0.1 - Expanding the vocabulary dictionary size from 32k to 62k to better support Traditional Chinese - 64k tokens context length - Multi-turn dialogue (without special handling for harmfulness) ## Model Details - Breeze-7B-Base-v0.1 - Finetuned from: [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) - Model type: Causal decoder-only transformer language model - Language: English and Traditional Chinese (zh-tw) - Breeze-7B-Instruct-v0.1 - Finetuned from: [MediaTek-Research/Breeze-7B-Base-v0.1](https://huggingface.co/MediaTek-Research/Breeze-7B-Base-v0.1) - Model type: Causal decoder-only transformer language model - Language: English and Traditional Chinese (zh-tw) - Breeze-7B-Instruct-64k-v0.1 - Finetuned from: [MediaTek-Research/Breeze-7B-Instruct-v0.1](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v0.1) - Model type: Causal decoder-only transformer language model - Language: English and Traditional Chinese (zh-tw) ## Base Model Performance **TMMLU+**, **DRCD**, and **Table** source from [MediaTek-Research/TCEval-v2](https://huggingface.co/datasets/MediaTek-Research/TCEval-v2). [MediaTek-Research/TCEval-v2](https://huggingface.co/datasets/MediaTek-Research/TCEval-v2) derives from [TCEval-v1](https://github.com/mtkresearch/MR-Models/tree/main/TC-Eval) and [ikala/tmmluplus](https://huggingface.co/datasets/ikala/tmmluplus). **MMLU** sources from [hails/mmlu_no_train](https://huggingface.co/datasets/hails/mmlu_no_train). We use the code revised from [EleutherAI/lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness) to evaluate **TMMLU+**, **DRCD**, **Table**, and **MMLU**. | Models | |↑ TMMLU+ (ACC) | DRCD (EM) | Table (ACC) | MMLU (ACC) | |----------------------------------------------|--------|--------------|-------------|-------------|------------| | | |TC, Knowledge |TC, Reasoning|TC, Reasoning|EN, Knowledge| | | | 5 shot | 3 shot | 5 shot | 5 shot | | [Yi-34B](https://huggingface.co/01-ai/Yi-34B)| 34B | 63.10 | 84.57 | 49.31 | 77.42 | | [Qwen-14B](https://huggingface.co/01-ai/Qwen/Qwen-14B)| 14B | 51.30 | 16.95 * | 50.69 | 68.83 | | [Yi-6B](https://huggingface.co/01-ai/Yi-6B) | 6B | 49.63 | 76.61 | 34.72 | 65.35 | | [Qwen-7B](https://huggingface.co/01-ai/Qwen/Qwen-7B)| 7B | 42.84 | 0.0 * | 39.58 | 61.00 | | [**Breeze-7B-Base-v0.1**](https://huggingface.co/MediaTek-Research/Breeze-7B-Base-v0.1) | 7B | 40.35 | 81.13 | 28.47 | 61.63 | | [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)| 7B | 36.93 | 79.27 | 27.78 | 64.89 | \* Few-shot learning cannot effectively guide the model to generate the proper answer. **Category ACC of TMMLU+ (5 shot)** | Models | STEM | Social Science | Humanities | Other | ↑ AVG | |----------------------------------|--------------|----------------|------------|------------|-------| | Yi-34B | 56.03 | 73.06 | 61.12 | 62.19 | 63.10 | | Qwen-14B | 46.51 | 58.20 | 51.12 | 49.38 | 51.30 | | Yi-6B | 41.14 | 57.77 | 50.22 | 49.39 | 49.63 | | Qwen-7B | 28.25 | 47.80 | 43.14 | 42.17 | 42.84 | | **Breeze-7B-Base-v0.1** | 35.74 | 46.08 | 40.29 | 39.27 | 40.35 | | Mistral-7B-v0.1 | 33.01 | 42.23 | 35.86 | 37.63 | 36.93 | ## Chat Model Performance **TMMLU+**, **DRCD**, **Table**, and **MT-Bench-tw** source from [MediaTek-Research/TCEval-v2](https://huggingface.co/datasets/MediaTek-Research/TCEval-v2). [MediaTek-Research/TCEval-v2](https://huggingface.co/datasets/MediaTek-Research/TCEval-v2) derives from [TCEval-v1](https://github.com/mtkresearch/MR-Models/tree/main/TC-Eval) and [ikala/tmmluplus](https://huggingface.co/datasets/ikala/tmmluplus). **MMLU** sources from [hails/mmlu_no_train](https://huggingface.co/datasets/hails/mmlu_no_train). **MT-Bench** source from [lmsys/mt_bench_human_judgments](https://huggingface.co/datasets/lmsys/mt_bench_human_judgments). We use the code revised from [EleutherAI/lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness) to evaluate **TMMLU+**, **DRCD**, **Table**, and **MMLU**. We use the code revised from [fastchat llm_judge](https://github.com/lm-sys/FastChat/tree/main/fastchat/llm_judge) (GPT4 as judge) to evaluate **MT-Bench-tw** and **MT-Bench**. | Models | |↑ MT-Bench-tw (Score)| TMMLU+ (ACC) | TMMLU+ (ACC) | DRCD (EM) | Table (ACC) | MT-Bench (Score) | MMLU (ACC) | MMLU (ACC) | |---------------------------------------------------------------------------------------------------------|--------|--------------------|--------------|--------------|-------------|-------------|------------------|-------------|-------------| | | |TC, Chat |TC, Knowledge |TC, Knowledge |TC, Reasoning|TC, Reasoning|EN, Chat |EN, Knowledge|EN, Knowledge| | | |0 shot | 0 shot | 5 shot | 3 shot | 0 shot |0 shot | 0 shot | 5 shot | | [gpt-3.5-turbo](https://openai.com) | |7.1 | 41.76 | | | |7.9 | 70.00 | | | [Yi-34B-Chat](https://huggingface.co/01-ai/Yi-34B-Chat) | 34B |6.9 | 54.87 | | | 36.81 |7.6 | 71.04 | | | [Qwen-14B-Chat](https://huggingface.co/Qwen/Qwen-14B-Chat) | 14B |6.4 | 48.41 | | | 41.67 |7.2 | 64.91 | | | [**Breeze-7B-Instruct-v0.1**](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v0.1) | 7B |5.7 | 41.61 | | | 45.83 |7.1 | 63.26 | | | [**Breeze-7B-Instruct-64k-v0.1**](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-64k-v0.1) | 7B |5.5 | 40.99 | | | 36.11 |7.1 | 63.68 | | | [Qwen-7B-Chat](https://huggingface.co/Qwen/Qwen-7B-Chat) | 7B |5.4 | 40.02 | | | 33.33 |6.2 | 55.94 | | | [Yi-6B-Chat](https://huggingface.co/01-ai/Yi-6B-Chat) | 6B |5.0 | 44.79 | | | 25.69 |6.0 | 59.45 | | | [Taiwan-LLM-13B-v2.0-chat](https://huggingface.co/yentinglin/Taiwan-LLM-13B-v2.0-chat) | 13B |5.0 | 29.47 | | | 23.61 |-* | 50.50 | | | [Taiwan-LLM-7B-v2.1-chat](https://huggingface.co/yentinglin/Taiwan-LLM-7B-v2.1-chat) | 7B |4.2 | 28.08 | | | 31.25 | -* | 42.72 | | \* Taiwan-LLM models responds to multi-turn questions (English) in Traditional Chinese. **Category Score of MT-Bench-tw (0 shot)** | Models | STEM |Extraction|Reasoning| Math | Coding | Roleplay| Writing |Humanities|↑ AVG | |-----------------------------------------------------|---------|---------|---------|---------|---------|---------|---------|---------|---------| | gpt-3.5-turbo | | | | | | | | | | | Yi-34B-Chat | | | | | | | | | | | Qwen-14B-Chat | | | | | | | | | | | **Breeze-7B-Instruct-v0.1** | | | | | | | | | | | **Breeze-7B-Instruct-64k-v0.1** | | | | | | | | | | | Qwen-7B-Chat | | | | | | | | | | | Yi-6B-Chat | | | | | | | | | | | Taiwan-LLM-13B-v2.0-chat | | | | | | | | | | | Taiwan-LLM-7B-v2.1-chat | | | | | | | | | | **Category ACC of TMMLU+ (0 shot)** | Model | STEM | Social Science | Humanities | Other | ↑ AVG | |-----------------------------------------------------|--------------|----------------|------------|------------|---------| | Yi-34B-Chat | 47.65 | 64.25 | 52.73 | 54.91 | 54.87 | | Qwen-14B-Chat | 43.83 | 55.00 | 48.55 | 46.22 | 48.41 | | Yi-6B-Chat | 37.80 | 51.74 | 45.36 | 44.25 | 44.79 | | gpt-3.5-turbo | 41.56 | 46.72 | 36.73 | 42.03 | 41.76 | | **Breeze-7B-Instruct-v0.1** | 37.41 | 46.81 | 42.06 | 40.16 | 41.61 | | **Breeze-7B-Instruct-64k-v0.1** | 37.88 | 46.35 | 40.31 | 39.40 | 40.99 | | Qwen-7B-Chat | 35.44 | 46.22 | 38.35 | 40.06 | 40.02 | | Taiwan-LLM-13B-v2.0-chat | 27.74 | 33.69 | 27.03 | 29.43 | 29.47 | | Taiwan-LLM-7B-v2.1-chat | 25.58 | 31.76 | 27.36 | 27.61 | 28.08 | ## Inference Performance In this test, we use the first 700 characters of the [web article](https://health.udn.com/health/story/5976/7699252?from=udn_ch1005_main_index) as the input and ask the model to write the same article again. All inferences run on 2 RTX A6000 GPUs (using `vllm`, with a tensor-parallel size of 2). | Models | ↓ Inference Time (sec)|Estimated Max Input Length (Char)| |--------------------------------------------------------------------|-------------------|--------------------------| | Yi-6B | 10.62 | 5.2k | | **Breeze-7B-Instruct-v0.1** | 10.74 | 11.1k | | **Breeze-7B-Instruct-64k-v0.1** | 10.74 | 88.8k | | Qwen-7B | 10.86 | 9.8k | | Qwen-14B | 18.89 | 9.8k | | Mistral-7B-v0.1 | 20.48 | 5.1k | | Taiwan-LLM-7B-v2.1-base | 26.26 | 2.2k | | Taiwan-LLM-13B-v2.0-base | 36.80 | 2.2k | | Yi-34B | 43.71 | 4.5k | ## Long-context Performance TBD ## Examples TBD ## Use in Transformers First install direct dependencies: ``` pip install transformers torch accelerate ``` If you want faster inference using flash-attention2, you need to install these dependencies: ```bash pip install packaging ninja pip install flash-attn ``` Then load the model in transformers: ```python from transformers import AutoModelForCausalLM, AutoTokenizer import torch model = AutoModelForCausalLM.from_pretrained( model="MediaTek-Research/Breeze-7B-Instruct-v0.1", device_map="auto", torch_dtype=torch.bfloat16, use_flash_attn_2=True # optional ) ``` The structure of the query template follows that of Mistral-7B-Instruct, as shown below. ```txt SYS_PROMPT [INST] QUERY1 [/INST] RESPONSE1 [INST] QUERY2 [/INST] ``` where `SYS_PROMPT`, `QUERY1`, `RESPONSE1`, and `QUERY2` can be provided by the user. The suggested default `SYS_PROMPT` is ```txt You are a helpful AI assistant built by MediaTek Research. The user you are helping speaks Traditional Chinese and comes from Taiwan. ``` ## Citation ``` @article{breeze7b2024, title={}, author={}, journal={arXiv}, year={2024} } ```