HuggingFaceH4/zephyr-7b-beta
1.9k60k
1---2tags:3- generated_from_trainer4license: mit5datasets:6- HuggingFaceH4/ultrachat_200k7- HuggingFaceH4/ultrafeedback_binarized8language:9- en10base_model: mistralai/Mistral-7B-v0.111widget:12 - example_title: Pirate!13 messages:14 - role: system15 content: You are a pirate chatbot who always responds with Arr!16 - role: user17 content: "There's a llama on my lawn, how can I get rid of him?"18 output:19 text: >-20 Arr! 'Tis a puzzlin' matter, me hearty! A llama on yer lawn be a rare21 sight, but I've got a plan that might help ye get rid of 'im. Ye'll need22 to gather some carrots and hay, and then lure the llama away with the23 promise of a tasty treat. Once he's gone, ye can clean up yer lawn and24 enjoy the peace and quiet once again. But beware, me hearty, for there25 may be more llamas where that one came from! Arr!26pipeline_tag: text-generation27model-index:28- name: zephyr-7b-beta29 results:30 # AI2 Reasoning Challenge (25-Shot)31 - task: 32 type: text-generation33 name: Text Generation34 dataset:35 name: AI2 Reasoning Challenge (25-Shot)36 type: ai2_arc37 config: ARC-Challenge38 split: test39 args:40 num_few_shot: 2541 metrics:42 - type: acc_norm43 name: normalized accuracy44 value: 62.0307167235494845 source:46 name: Open LLM Leaderboard47 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta48 49 # HellaSwag (10-shot)50 - task: 51 type: text-generation52 name: Text Generation53 dataset:54 name: HellaSwag (10-Shot)55 type: hellaswag56 split: validation57 args:58 num_few_shot: 1059 metrics:60 - type: acc_norm61 name: normalized accuracy62 value: 84.3557060346544563 source:64 name: Open LLM Leaderboard65 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta66 67 # DROP (3-shot)68 - task: 69 type: text-generation70 name: Text Generation71 dataset:72 name: Drop (3-Shot)73 type: drop74 split: validation75 args:76 num_few_shot: 377 metrics:78 - type: f179 name: f1 score80 value: 9.66243708053690981 source:82 name: Open LLM Leaderboard83 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta84 85 # TruthfulQA (0-shot)86 - task: 87 type: text-generation88 name: Text Generation89 dataset:90 name: TruthfulQA (0-shot)91 type: truthful_qa92 config: multiple_choice93 split: validation94 args:95 num_few_shot: 096 metrics:97 - type: mc298 value: 57.4491694276285599 source:100 name: Open LLM Leaderboard101 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta102 103 # GSM8k (5-shot)104 - task: 105 type: text-generation106 name: Text Generation107 dataset:108 name: GSM8k (5-shot)109 type: gsm8k110 config: main111 split: test112 args:113 num_few_shot: 5114 metrics:115 - type: acc116 name: accuracy117 value: 12.736921910538287118 source:119 name: Open LLM Leaderboard120 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta121 122 # MMLU (5-Shot)123 - task: 124 type: text-generation125 name: Text Generation126 dataset:127 name: MMLU (5-Shot)128 type: cais/mmlu129 config: all130 split: test131 args:132 num_few_shot: 5133 metrics:134 - type: acc135 name: accuracy136 value: 61.07137 source:138 name: Open LLM Leaderboard139 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta140 141 # Winogrande (5-shot)142 - task: 143 type: text-generation144 name: Text Generation145 dataset:146 name: Winogrande (5-shot)147 type: winogrande148 config: winogrande_xl149 split: validation150 args:151 num_few_shot: 5152 metrics:153 - type: acc154 name: accuracy155 value: 77.74269928966061156 source:157 name: Open LLM Leaderboard158 url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HuggingFaceH4/zephyr-7b-beta159 160 # AlpacaEval (taken from model card)161 - task: 162 type: text-generation163 name: Text Generation164 dataset:165 name: AlpacaEval166 type: tatsu-lab/alpaca_eval167 metrics:168 - type: unknown169 name: win rate170 value: 0.9060171 source:172 url: https://tatsu-lab.github.io/alpaca_eval/173 174 # MT-Bench (taken from model card)175 - task: 176 type: text-generation177 name: Text Generation178 dataset:179 name: MT-Bench180 type: unknown181 metrics:182 - type: unknown183 name: score184 value: 7.34185 source:186 url: https://huggingface.co/spaces/lmsys/mt-bench187---188 189<!-- This model card has been generated automatically according to the information the Trainer had access to. You190should probably proofread and complete it, then remove this comment. -->191 192<img src="https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha/resolve/main/thumbnail.png" alt="Zephyr Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>193 194 195# Model Card for Zephyr 7B β196 197Zephyr is a series of language models that are trained to act as helpful assistants. Zephyr-7B-β is the second model in the series, and is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) that was trained on on a mix of publicly available, synthetic datasets using [Direct Preference Optimization (DPO)](https://arxiv.org/abs/2305.18290). We found that removing the in-built alignment of these datasets boosted performance on [MT Bench](https://huggingface.co/spaces/lmsys/mt-bench) and made the model more helpful. However, this means that model is likely to generate problematic text when prompted to do so. You can find more details in the [technical report](https://arxiv.org/abs/2310.16944).198 199 200## Model description201 202- **Model type:** A 7B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.203- **Language(s) (NLP):** Primarily English204- **License:** MIT205- **Finetuned from model:** [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)206 207### Model Sources208 209<!-- Provide the basic links for the model. -->210 211- **Repository:** https://github.com/huggingface/alignment-handbook212- **Demo:** https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat213- **Chatbot Arena:** Evaluate Zephyr 7B against 10+ LLMs in the LMSYS arena: http://arena.lmsys.org214 215## Performance216 217At the time of release, Zephyr-7B-β is the highest ranked 7B chat model on the [MT-Bench](https://huggingface.co/spaces/lmsys/mt-bench) and [AlpacaEval](https://tatsu-lab.github.io/alpaca_eval/) benchmarks:218 219| Model | Size | Alignment | MT-Bench (score) | AlpacaEval (win rate %) |220|-------------|-----|----|---------------|--------------|221| StableLM-Tuned-α | 7B| dSFT |2.75| -|222| MPT-Chat | 7B |dSFT |5.42| -|223| Xwin-LMv0.1 | 7B| dPPO| 6.19| 87.83|224| Mistral-Instructv0.1 | 7B| - | 6.84 |-|225| Zephyr-7b-α |7B| dDPO| 6.88| -|226| **Zephyr-7b-β** 🪁 | **7B** | **dDPO** | **7.34** | **90.60** |227| Falcon-Instruct | 40B |dSFT |5.17 |45.71|228| Guanaco | 65B | SFT |6.41| 71.80|229| Llama2-Chat | 70B |RLHF |6.86| 92.66|230| Vicuna v1.3 | 33B |dSFT |7.12 |88.99|231| WizardLM v1.0 | 70B |dSFT |7.71 |-|232| Xwin-LM v0.1 | 70B |dPPO |- |95.57|233| GPT-3.5-turbo | - |RLHF |7.94 |89.37|234| Claude 2 | - |RLHF |8.06| 91.36|235| GPT-4 | -| RLHF |8.99| 95.28|236 237In particular, on several categories of MT-Bench, Zephyr-7B-β has strong performance compared to larger open models like Llama2-Chat-70B:238 239240 241However, on more complex tasks like coding and mathematics, Zephyr-7B-β lags behind proprietary models and more research is needed to close the gap.242 243 244## Intended uses & limitations245 246The model was initially fine-tuned on a filtered and preprocessed of the [`UltraChat`](https://huggingface.co/datasets/stingning/ultrachat) dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT. 247We then further aligned the model with [🤗 TRL's](https://github.com/huggingface/trl) `DPOTrainer` on the [openbmb/UltraFeedback](https://huggingface.co/datasets/openbmb/UltraFeedback) dataset, which contains 64k prompts and model completions that are ranked by GPT-4. As a result, the model can be used for chat and you can check out our [demo](https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat) to test its capabilities. 248 249You can find the datasets used for training Zephyr-7B-β [here](https://huggingface.co/collections/HuggingFaceH4/zephyr-7b-6538c6d6d5ddd1cbb1744a66)250 251Here's how you can run the model using the `pipeline()` function from 🤗 Transformers:252 253```python254# Install transformers from source - only needed for versions <= v4.34255# pip install git+https://github.com/huggingface/transformers.git256# pip install accelerate257 258import torch259from transformers import pipeline260 261pipe = pipeline("text-generation", model="HuggingFaceH4/zephyr-7b-beta", torch_dtype=torch.bfloat16, device_map="auto")262 263# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating264messages = [265 {266 "role": "system",267 "content": "You are a friendly chatbot who always responds in the style of a pirate",268 },269 {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},270]271prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)272outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)273print(outputs[0]["generated_text"])274# <|system|>275# You are a friendly chatbot who always responds in the style of a pirate.</s>276# <|user|>277# How many helicopters can a human eat in one sitting?</s>278# <|assistant|>279# Ah, me hearty matey! But yer question be a puzzler! A human cannot eat a helicopter in one sitting, as helicopters are not edible. They be made of metal, plastic, and other materials, not food!280```281 282## Bias, Risks, and Limitations283 284<!-- This section is meant to convey both technical and sociotechnical limitations. -->285 286Zephyr-7B-β has not been aligned to human preferences for safety within the RLHF phase or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so). 287It is also unknown what the size and composition of the corpus was used to train the base model (`mistralai/Mistral-7B-v0.1`), however it is likely to have included a mix of Web data and technical sources like books and code. See the [Falcon 180B model card](https://huggingface.co/tiiuae/falcon-180B#training-data) for an example of this.288 289 290## Training and evaluation data291 292During DPO training, this model achieves the following results on the evaluation set:293 294- Loss: 0.7496295- Rewards/chosen: -4.5221296- Rewards/rejected: -8.3184297- Rewards/accuracies: 0.7812298- Rewards/margins: 3.7963299- Logps/rejected: -340.1541300- Logps/chosen: -299.4561301- Logits/rejected: -2.3081302- Logits/chosen: -2.3531303 304 305### Training hyperparameters306 307The following hyperparameters were used during training:308- learning_rate: 5e-07309- train_batch_size: 2310- eval_batch_size: 4311- seed: 42312- distributed_type: multi-GPU313- num_devices: 16314- total_train_batch_size: 32315- total_eval_batch_size: 64316- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08317- lr_scheduler_type: linear318- lr_scheduler_warmup_ratio: 0.1319- num_epochs: 3.0320 321### Training results322 323The table below shows the full set of DPO training metrics:324 325 326| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |327|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|328| 0.6284 | 0.05 | 100 | 0.6098 | 0.0425 | -0.1872 | 0.7344 | 0.2297 | -258.8416 | -253.8099 | -2.7976 | -2.8234 |329| 0.4908 | 0.1 | 200 | 0.5426 | -0.0279 | -0.6842 | 0.75 | 0.6563 | -263.8124 | -254.5145 | -2.7719 | -2.7960 |330| 0.5264 | 0.15 | 300 | 0.5324 | 0.0414 | -0.9793 | 0.7656 | 1.0207 | -266.7627 | -253.8209 | -2.7892 | -2.8122 |331| 0.5536 | 0.21 | 400 | 0.4957 | -0.0185 | -1.5276 | 0.7969 | 1.5091 | -272.2460 | -254.4203 | -2.8542 | -2.8764 |332| 0.5362 | 0.26 | 500 | 0.5031 | -0.2630 | -1.5917 | 0.7812 | 1.3287 | -272.8869 | -256.8653 | -2.8702 | -2.8958 |333| 0.5966 | 0.31 | 600 | 0.5963 | -0.2993 | -1.6491 | 0.7812 | 1.3499 | -273.4614 | -257.2279 | -2.8778 | -2.8986 |334| 0.5014 | 0.36 | 700 | 0.5382 | -0.2859 | -1.4750 | 0.75 | 1.1891 | -271.7204 | -257.0942 | -2.7659 | -2.7869 |335| 0.5334 | 0.41 | 800 | 0.5677 | -0.4289 | -1.8968 | 0.7969 | 1.4679 | -275.9378 | -258.5242 | -2.7053 | -2.7265 |336| 0.5251 | 0.46 | 900 | 0.5772 | -0.2116 | -1.3107 | 0.7344 | 1.0991 | -270.0768 | -256.3507 | -2.8463 | -2.8662 |337| 0.5205 | 0.52 | 1000 | 0.5262 | -0.3792 | -1.8585 | 0.7188 | 1.4793 | -275.5552 | -258.0276 | -2.7893 | -2.7979 |338| 0.5094 | 0.57 | 1100 | 0.5433 | -0.6279 | -1.9368 | 0.7969 | 1.3089 | -276.3377 | -260.5136 | -2.7453 | -2.7536 |339| 0.5837 | 0.62 | 1200 | 0.5349 | -0.3780 | -1.9584 | 0.7656 | 1.5804 | -276.5542 | -258.0154 | -2.7643 | -2.7756 |340| 0.5214 | 0.67 | 1300 | 0.5732 | -1.0055 | -2.2306 | 0.7656 | 1.2251 | -279.2761 | -264.2903 | -2.6986 | -2.7113 |341| 0.6914 | 0.72 | 1400 | 0.5137 | -0.6912 | -2.1775 | 0.7969 | 1.4863 | -278.7448 | -261.1467 | -2.7166 | -2.7275 |342| 0.4655 | 0.77 | 1500 | 0.5090 | -0.7987 | -2.2930 | 0.7031 | 1.4943 | -279.8999 | -262.2220 | -2.6651 | -2.6838 |343| 0.5731 | 0.83 | 1600 | 0.5312 | -0.8253 | -2.3520 | 0.7812 | 1.5268 | -280.4902 | -262.4876 | -2.6543 | -2.6728 |344| 0.5233 | 0.88 | 1700 | 0.5206 | -0.4573 | -2.0951 | 0.7812 | 1.6377 | -277.9205 | -258.8084 | -2.6870 | -2.7097 |345| 0.5593 | 0.93 | 1800 | 0.5231 | -0.5508 | -2.2000 | 0.7969 | 1.6492 | -278.9703 | -259.7433 | -2.6221 | -2.6519 |346| 0.4967 | 0.98 | 1900 | 0.5290 | -0.5340 | -1.9570 | 0.8281 | 1.4230 | -276.5395 | -259.5749 | -2.6564 | -2.6878 |347| 0.0921 | 1.03 | 2000 | 0.5368 | -1.1376 | -3.1615 | 0.7812 | 2.0239 | -288.5854 | -265.6111 | -2.6040 | -2.6345 |348| 0.0733 | 1.08 | 2100 | 0.5453 | -1.1045 | -3.4451 | 0.7656 | 2.3406 | -291.4208 | -265.2799 | -2.6289 | -2.6595 |349| 0.0972 | 1.14 | 2200 | 0.5571 | -1.6915 | -3.9823 | 0.8125 | 2.2908 | -296.7934 | -271.1505 | -2.6471 | -2.6709 |350| 0.1058 | 1.19 | 2300 | 0.5789 | -1.0621 | -3.8941 | 0.7969 | 2.8319 | -295.9106 | -264.8563 | -2.5527 | -2.5798 |351| 0.2423 | 1.24 | 2400 | 0.5455 | -1.1963 | -3.5590 | 0.7812 | 2.3627 | -292.5599 | -266.1981 | -2.5414 | -2.5784 |352| 0.1177 | 1.29 | 2500 | 0.5889 | -1.8141 | -4.3942 | 0.7969 | 2.5801 | -300.9120 | -272.3761 | -2.4802 | -2.5189 |353| 0.1213 | 1.34 | 2600 | 0.5683 | -1.4608 | -3.8420 | 0.8125 | 2.3812 | -295.3901 | -268.8436 | -2.4774 | -2.5207 |354| 0.0889 | 1.39 | 2700 | 0.5890 | -1.6007 | -3.7337 | 0.7812 | 2.1330 | -294.3068 | -270.2423 | -2.4123 | -2.4522 |355| 0.0995 | 1.45 | 2800 | 0.6073 | -1.5519 | -3.8362 | 0.8281 | 2.2843 | -295.3315 | -269.7538 | -2.4685 | -2.5050 |356| 0.1145 | 1.5 | 2900 | 0.5790 | -1.7939 | -4.2876 | 0.8438 | 2.4937 | -299.8461 | -272.1744 | -2.4272 | -2.4674 |357| 0.0644 | 1.55 | 3000 | 0.5735 | -1.7285 | -4.2051 | 0.8125 | 2.4766 | -299.0209 | -271.5201 | -2.4193 | -2.4574 |358| 0.0798 | 1.6 | 3100 | 0.5537 | -1.7226 | -4.2850 | 0.8438 | 2.5624 | -299.8200 | -271.4610 | -2.5367 | -2.5696 |359| 0.1013 | 1.65 | 3200 | 0.5575 | -1.5715 | -3.9813 | 0.875 | 2.4098 | -296.7825 | -269.9498 | -2.4926 | -2.5267 |360| 0.1254 | 1.7 | 3300 | 0.5905 | -1.6412 | -4.4703 | 0.8594 | 2.8291 | -301.6730 | -270.6473 | -2.5017 | -2.5340 |361| 0.085 | 1.76 | 3400 | 0.6133 | -1.9159 | -4.6760 | 0.8438 | 2.7601 | -303.7296 | -273.3941 | -2.4614 | -2.4960 |362| 0.065 | 1.81 | 3500 | 0.6074 | -1.8237 | -4.3525 | 0.8594 | 2.5288 | -300.4951 | -272.4724 | -2.4597 | -2.5004 |363| 0.0755 | 1.86 | 3600 | 0.5836 | -1.9252 | -4.4005 | 0.8125 | 2.4753 | -300.9748 | -273.4872 | -2.4327 | -2.4716 |364| 0.0746 | 1.91 | 3700 | 0.5789 | -1.9280 | -4.4906 | 0.8125 | 2.5626 | -301.8762 | -273.5149 | -2.4686 | -2.5115 |365| 0.1348 | 1.96 | 3800 | 0.6015 | -1.8658 | -4.2428 | 0.8281 | 2.3769 | -299.3976 | -272.8936 | -2.4943 | -2.5393 |366| 0.0217 | 2.01 | 3900 | 0.6122 | -2.3335 | -4.9229 | 0.8281 | 2.5894 | -306.1988 | -277.5699 | -2.4841 | -2.5272 |367| 0.0219 | 2.07 | 4000 | 0.6522 | -2.9890 | -6.0164 | 0.8281 | 3.0274 | -317.1334 | -284.1248 | -2.4105 | -2.4545 |368| 0.0119 | 2.12 | 4100 | 0.6922 | -3.4777 | -6.6749 | 0.7969 | 3.1972 | -323.7187 | -289.0121 | -2.4272 | -2.4699 |369| 0.0153 | 2.17 | 4200 | 0.6993 | -3.2406 | -6.6775 | 0.7969 | 3.4369 | -323.7453 | -286.6413 | -2.4047 | -2.4465 |370| 0.011 | 2.22 | 4300 | 0.7178 | -3.7991 | -7.4397 | 0.7656 | 3.6406 | -331.3667 | -292.2260 | -2.3843 | -2.4290 |371| 0.0072 | 2.27 | 4400 | 0.6840 | -3.3269 | -6.8021 | 0.8125 | 3.4752 | -324.9908 | -287.5042 | -2.4095 | -2.4536 |372| 0.0197 | 2.32 | 4500 | 0.7013 | -3.6890 | -7.3014 | 0.8125 | 3.6124 | -329.9841 | -291.1250 | -2.4118 | -2.4543 |373| 0.0182 | 2.37 | 4600 | 0.7476 | -3.8994 | -7.5366 | 0.8281 | 3.6372 | -332.3356 | -293.2291 | -2.4163 | -2.4565 |374| 0.0125 | 2.43 | 4700 | 0.7199 | -4.0560 | -7.5765 | 0.8438 | 3.5204 | -332.7345 | -294.7952 | -2.3699 | -2.4100 |375| 0.0082 | 2.48 | 4800 | 0.7048 | -3.6613 | -7.1356 | 0.875 | 3.4743 | -328.3255 | -290.8477 | -2.3925 | -2.4303 |376| 0.0118 | 2.53 | 4900 | 0.6976 | -3.7908 | -7.3152 | 0.8125 | 3.5244 | -330.1224 | -292.1431 | -2.3633 | -2.4047 |377| 0.0118 | 2.58 | 5000 | 0.7198 | -3.9049 | -7.5557 | 0.8281 | 3.6508 | -332.5271 | -293.2844 | -2.3764 | -2.4194 |378| 0.006 | 2.63 | 5100 | 0.7506 | -4.2118 | -7.9149 | 0.8125 | 3.7032 | -336.1194 | -296.3530 | -2.3407 | -2.3860 |379| 0.0143 | 2.68 | 5200 | 0.7408 | -4.2433 | -7.9802 | 0.8125 | 3.7369 | -336.7721 | -296.6682 | -2.3509 | -2.3946 |380| 0.0057 | 2.74 | 5300 | 0.7552 | -4.3392 | -8.0831 | 0.7969 | 3.7439 | -337.8013 | -297.6275 | -2.3388 | -2.3842 |381| 0.0138 | 2.79 | 5400 | 0.7404 | -4.2395 | -7.9762 | 0.8125 | 3.7367 | -336.7322 | -296.6304 | -2.3286 | -2.3737 |382| 0.0079 | 2.84 | 5500 | 0.7525 | -4.4466 | -8.2196 | 0.7812 | 3.7731 | -339.1662 | -298.7007 | -2.3200 | -2.3641 |383| 0.0077 | 2.89 | 5600 | 0.7520 | -4.5586 | -8.3485 | 0.7969 | 3.7899 | -340.4545 | -299.8206 | -2.3078 | -2.3517 |384| 0.0094 | 2.94 | 5700 | 0.7527 | -4.5542 | -8.3509 | 0.7812 | 3.7967 | -340.4790 | -299.7773 | -2.3062 | -2.3510 |385| 0.0054 | 2.99 | 5800 | 0.7520 | -4.5169 | -8.3079 | 0.7812 | 3.7911 | -340.0493 | -299.4038 | -2.3081 | -2.3530 |386 387 388### Framework versions389 390- Transformers 4.35.0.dev0391- Pytorch 2.0.1+cu118392- Datasets 2.12.0393- Tokenizers 0.14.0394 395## Citation396 397If you find Zephyr-7B-β is useful in your work, please cite it with:398 399```400@misc{tunstall2023zephyr,401 title={Zephyr: Direct Distillation of LM Alignment}, 402 author={Lewis Tunstall and Edward Beeching and Nathan Lambert and Nazneen Rajani and Kashif Rasul and Younes Belkada and Shengyi Huang and Leandro von Werra and Clémentine Fourrier and Nathan Habib and Nathan Sarrazin and Omar Sanseviero and Alexander M. Rush and Thomas Wolf},403 year={2023},404 eprint={2310.16944},405 archivePrefix={arXiv},406 primaryClass={cs.LG}407}408```409 410If you use the UltraChat or UltraFeedback datasets, please cite the original works:411 412```413@misc{ding2023enhancing,414 title={Enhancing Chat Language Models by Scaling High-quality Instructional Conversations}, 415 author={Ning Ding and Yulin Chen and Bokai Xu and Yujia Qin and Zhi Zheng and Shengding Hu and Zhiyuan Liu and Maosong Sun and Bowen Zhou},416 year={2023},417 eprint={2305.14233},418 archivePrefix={arXiv},419 primaryClass={cs.CL}420}421 422@misc{cui2023ultrafeedback,423 title={UltraFeedback: Boosting Language Models with High-quality Feedback}, 424 author={Ganqu Cui and Lifan Yuan and Ning Ding and Guanming Yao and Wei Zhu and Yuan Ni and Guotong Xie and Zhiyuan Liu and Maosong Sun},425 year={2023},426 eprint={2310.01377},427 archivePrefix={arXiv},428 primaryClass={cs.CL}429}430```431 432# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)433Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_HuggingFaceH4__zephyr-7b-beta)434 435| Metric | Value |436|-----------------------|---------------------------|437| Avg. | 52.15 |438| ARC (25-shot) | 62.03 |439| HellaSwag (10-shot) | 84.36 |440| MMLU (5-shot) | 61.07 |441| TruthfulQA (0-shot) | 57.45 |442| Winogrande (5-shot) | 77.74 |443| GSM8K (5-shot) | 12.74 |444| DROP (3-shot) | 9.66 |