datasets:
- Trelis/openassistant-yi
language:
- en
inference: false
tags:
- yi
- long context
- commercial use
- gguf
- awq
extra_gated_prompt: >-
Access to this repo requires the purchase of a license (see link on model card
below)
extra_gated_fields:
Name: text
Affiliation: text
I have purchased access (access will be granted once your payment clears): checkbox
I agree to the terms of the license described on the dataset card: checkbox
✨ Yi 200k context SFT models
These are chat fine-tuned versions of the Yi 200k context length models:
- Supervised Fine-tuning allows the model to respond in a cleaner chat format that ends with EOS tokens.
- Note that this is a fine-tune of the llamafied model, meaning that all llama platforms can be used for inference.
Available models:
GGUF models are in the base model repos (along with the bf16 weight safetensors). AWQ models are in the '-AWQ' repos (34B AWQ will be released by EOD 20 Nov 2023). When you purchase access, you get access to all model variants for that model size.
Notably:
- The data used for fine-tuning is Apache 2 licensed and not generated using AI, thereby allowing this chat model to be used commercially, which is particularly useful for data preparation and generation for training other models.
- The purchase of access to this model grants the user permission to use the model commercially for inference or fine-tuning and inference.
Prompt format:
# Yi style
B_INST, E_INST = "Human: ", " Assistant:"
prompt = f"{B_INST}{user_prompt.strip()}{E_INST}"
THE ORIGINAL MODEL CARD FOLLOWS BELOW.
Llamafied version of 01-ai's Yi-6B-200k for ease of use.
Model Performance
| Model | MMLU | CMMLU | C-Eval | GAOKAO | BBH | Common-sense Reasoning | Reading Comprehension | Math & Code |
|---|---|---|---|---|---|---|---|---|
| 5-shot | 5-shot | 5-shot | 0-shot | 3-shot@1 | - | - | - | |
| LLaMA2-34B | 62.6 | - | - | - | 44.1 | 69.9 | 68.0 | 26.0 |
| LLaMA2-70B | 68.9 | 53.3 | - | 49.8 | 51.2 | 71.9 | 69.4 | 36.8 |
| Baichuan2-13B | 59.2 | 62.0 | 58.1 | 54.3 | 48.8 | 64.3 | 62.4 | 23.0 |
| Qwen-14B | 66.3 | 71.0 | 72.1 | 62.5 | 53.4 | 73.3 | 72.5 | 39.8 |
| Skywork-13B | 62.1 | 61.8 | 60.6 | 68.1 | 41.7 | 72.4 | 61.4 | 24.9 |
| InternLM-20B | 62.1 | 59.0 | 58.8 | 45.5 | 52.5 | 78.3 | - | 30.4 |
| Aquila-34B | 67.8 | 71.4 | 63.1 | - | - | - | - | - |
| Falcon-180B | 70.4 | 58.0 | 57.8 | 59.0 | 54.0 | 77.3 | 68.8 | 34.0 |
| Yi-6B | 63.2 | 75.5 | 72.0 | 72.2 | 42.8 | 72.3 | 68.7 | 19.8 |
| Yi-6B-200K | 64.0 | 75.3 | 73.5 | 73.9 | 42.0 | 72.0 | 69.1 | 19.0 |
| Yi-34B | 76.3 | 83.7 | 81.4 | 82.8 | 54.3 | 80.1 | 76.4 | 37.1 |
| Yi-34B-200K | 76.1 | 83.6 | 81.9 | 83.4 | 52.7 | 79.7 | 76.6 | 36.3 |
While benchmarking open-source models, we have observed a disparity between the results generated by our pipeline and those reported in public sources (e.g. OpenCompass). Upon conducting a more in-depth investigation of this difference, we have discovered that various models may employ different prompts, post-processing strategies, and sampling techniques, potentially resulting in significant variations in the outcomes. Our prompt and post-processing strategy remains consistent with the original benchmark, and greedy decoding is employed during evaluation without any post-processing for the generated content. For scores that were not reported by the original authors (including scores reported with different settings), we try to get results with our pipeline.
To evaluate the model's capability extensively, we adopted the methodology outlined in Llama2. Specifically, we included PIQA, SIQA, HellaSwag, WinoGrande, ARC, OBQA, and CSQA to assess common sense reasoning. SquAD, QuAC, and BoolQ were incorporated to evaluate reading comprehension. CSQA was exclusively tested using a 7-shot setup, while all other tests were conducted with a 0-shot configuration. Additionally, we introduced GSM8K (8-shot@1), MATH (4-shot@1), HumanEval (0-shot@1), and MBPP (3-shot@1) under the category "Math & Code". Due to technical constraints, we did not test Falcon-180 on QuAC and OBQA; the score is derived by averaging the scores on the remaining tasks. Since the scores for these two tasks are generally lower than the average, we believe that Falcon-180B's performance was not underestimated.
Usage
Please visit our github repository for general guidance on how to use this model.
Disclaimer
Although we use data compliance checking algorithms during the training process to ensure the compliance of the trained model to the best of our ability, due to the complexity of the data and the diversity of language model usage scenarios, we cannot guarantee that the model will generate correct and reasonable output in all scenarios. Please be aware that there is still a risk of the model producing problematic outputs. We will not be responsible for any risks and issues resulting from misuse, misguidance, illegal usage, and related misinformation, as well as any associated data security concerns.
License
The Yi series models are fully open for academic research and free commercial usage with permission via applications. All usage must adhere to the Model License Agreement 2.0. To apply for the official commercial license, please contact us (yi@01.ai).