Commit
•
5f13ded
1
Parent(s):
77e885e
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,142 +1 @@
|
|
| 1 |
-
|
| 2 |
-
tags:
|
| 3 |
-
- generated_from_trainer
|
| 4 |
-
model-index:
|
| 5 |
-
- name: zephyr-7b-alpha
|
| 6 |
-
results: []
|
| 7 |
-
license: mit
|
| 8 |
-
datasets:
|
| 9 |
-
- stingning/ultrachat
|
| 10 |
-
- openbmb/UltraFeedback
|
| 11 |
-
language:
|
| 12 |
-
- en
|
| 13 |
-
---
|
| 14 |
-
|
| 15 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
-
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
-
|
| 18 |
-
<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'"/>
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
# Model Card for Zephyr 7B Alpha
|
| 22 |
-
|
| 23 |
-
Zephyr is a series of language models that are trained to act as helpful assistants. Zephyr-7B-α is the first 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 and should only be used for educational and research purposes.
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
## Model description
|
| 27 |
-
|
| 28 |
-
- **Model type:** A 7B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
|
| 29 |
-
- **Language(s) (NLP):** Primarily English
|
| 30 |
-
- **License:** MIT
|
| 31 |
-
- **Finetuned from model:** [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)
|
| 32 |
-
|
| 33 |
-
### Model Sources
|
| 34 |
-
|
| 35 |
-
<!-- Provide the basic links for the model. -->
|
| 36 |
-
|
| 37 |
-
- **Repository:** https://github.com/huggingface/alignment-handbook
|
| 38 |
-
- **Demo:** https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat
|
| 39 |
-
|
| 40 |
-
## Intended uses & limitations
|
| 41 |
-
|
| 42 |
-
The model was initially fine-tuned on a variant of the [`UltraChat`](https://huggingface.co/datasets/stingning/ultrachat) dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT. We 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 contain 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.
|
| 43 |
-
|
| 44 |
-
Here's how you can run the model using the `pipeline()` function from 🤗 Transformers:
|
| 45 |
-
|
| 46 |
-
```python
|
| 47 |
-
import torch
|
| 48 |
-
from transformers import pipeline
|
| 49 |
-
|
| 50 |
-
pipe = pipeline("text-generation", model="HuggingFaceH4/zephyr-7b-alpha", torch_dtype=torch.bfloat16, device_map="auto")
|
| 51 |
-
|
| 52 |
-
# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
|
| 53 |
-
messages = [
|
| 54 |
-
{
|
| 55 |
-
"role": "system",
|
| 56 |
-
"content": "You are a friendly chatbot who always responds in the style of a pirate",
|
| 57 |
-
},
|
| 58 |
-
{"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
|
| 59 |
-
]
|
| 60 |
-
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
| 61 |
-
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 62 |
-
print(outputs[0]["generated_text"])
|
| 63 |
-
# <|system|>
|
| 64 |
-
# You are a friendly chatbot who always responds in the style of a pirate.</s>
|
| 65 |
-
# <|user|>
|
| 66 |
-
# How many helicopters can a human eat in one sitting?</s>
|
| 67 |
-
# <|assistant|>
|
| 68 |
-
# 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!
|
| 69 |
-
```
|
| 70 |
-
|
| 71 |
-
## Bias, Risks, and Limitations
|
| 72 |
-
|
| 73 |
-
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 74 |
-
|
| 75 |
-
Zephyr-7B-α has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so).
|
| 76 |
-
It 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.
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
## Training and evaluation data
|
| 80 |
-
|
| 81 |
-
Zephyr 7B Alpha achieves the following results on the evaluation set:
|
| 82 |
-
|
| 83 |
-
- Loss: 0.4605
|
| 84 |
-
- Rewards/chosen: -0.5053
|
| 85 |
-
- Rewards/rejected: -1.8752
|
| 86 |
-
- Rewards/accuracies: 0.7812
|
| 87 |
-
- Rewards/margins: 1.3699
|
| 88 |
-
- Logps/rejected: -327.4286
|
| 89 |
-
- Logps/chosen: -297.1040
|
| 90 |
-
- Logits/rejected: -2.7153
|
| 91 |
-
- Logits/chosen: -2.7447
|
| 92 |
-
|
| 93 |
-
## Training procedure
|
| 94 |
-
|
| 95 |
-
### Training hyperparameters
|
| 96 |
-
|
| 97 |
-
The following hyperparameters were used during training:
|
| 98 |
-
|
| 99 |
-
- learning_rate: 5e-07
|
| 100 |
-
- train_batch_size: 2
|
| 101 |
-
- eval_batch_size: 4
|
| 102 |
-
- seed: 42
|
| 103 |
-
- distributed_type: multi-GPU
|
| 104 |
-
- num_devices: 16
|
| 105 |
-
- total_train_batch_size: 32
|
| 106 |
-
- total_eval_batch_size: 64
|
| 107 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 108 |
-
- lr_scheduler_type: linear
|
| 109 |
-
- lr_scheduler_warmup_ratio: 0.1
|
| 110 |
-
- num_epochs: 1
|
| 111 |
-
|
| 112 |
-
### Training results
|
| 113 |
-
|
| 114 |
-
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
| 115 |
-
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
| 116 |
-
| 0.5602 | 0.05 | 100 | 0.5589 | -0.3359 | -0.8168 | 0.7188 | 0.4809 | -306.2607 | -293.7161 | -2.6554 | -2.6797 |
|
| 117 |
-
| 0.4852 | 0.1 | 200 | 0.5136 | -0.5310 | -1.4994 | 0.8125 | 0.9684 | -319.9124 | -297.6181 | -2.5762 | -2.5957 |
|
| 118 |
-
| 0.5212 | 0.15 | 300 | 0.5168 | -0.1686 | -1.1760 | 0.7812 | 1.0074 | -313.4444 | -290.3699 | -2.6865 | -2.7125 |
|
| 119 |
-
| 0.5496 | 0.21 | 400 | 0.4835 | -0.1617 | -1.7170 | 0.8281 | 1.5552 | -324.2635 | -290.2326 | -2.7947 | -2.8218 |
|
| 120 |
-
| 0.5209 | 0.26 | 500 | 0.5054 | -0.4778 | -1.6604 | 0.7344 | 1.1826 | -323.1325 | -296.5546 | -2.8388 | -2.8667 |
|
| 121 |
-
| 0.4617 | 0.31 | 600 | 0.4910 | -0.3738 | -1.5180 | 0.7656 | 1.1442 | -320.2848 | -294.4741 | -2.8234 | -2.8521 |
|
| 122 |
-
| 0.4452 | 0.36 | 700 | 0.4838 | -0.4591 | -1.6576 | 0.7031 | 1.1986 | -323.0770 | -296.1796 | -2.7401 | -2.7653 |
|
| 123 |
-
| 0.4674 | 0.41 | 800 | 0.5077 | -0.5692 | -1.8659 | 0.7656 | 1.2967 | -327.2416 | -298.3818 | -2.6740 | -2.6945 |
|
| 124 |
-
| 0.4656 | 0.46 | 900 | 0.4927 | -0.5279 | -1.6614 | 0.7656 | 1.1335 | -323.1518 | -297.5553 | -2.7817 | -2.8015 |
|
| 125 |
-
| 0.4102 | 0.52 | 1000 | 0.4772 | -0.5767 | -2.0667 | 0.7656 | 1.4900 | -331.2578 | -298.5311 | -2.7160 | -2.7455 |
|
| 126 |
-
| 0.4663 | 0.57 | 1100 | 0.4740 | -0.8038 | -2.1018 | 0.7656 | 1.2980 | -331.9604 | -303.0741 | -2.6994 | -2.7257 |
|
| 127 |
-
| 0.4737 | 0.62 | 1200 | 0.4716 | -0.3783 | -1.7015 | 0.7969 | 1.3232 | -323.9545 | -294.5634 | -2.6842 | -2.7135 |
|
| 128 |
-
| 0.4259 | 0.67 | 1300 | 0.4866 | -0.6239 | -1.9703 | 0.7812 | 1.3464 | -329.3312 | -299.4761 | -2.7046 | -2.7356 |
|
| 129 |
-
| 0.4935 | 0.72 | 1400 | 0.4747 | -0.5626 | -1.7600 | 0.7812 | 1.1974 | -325.1243 | -298.2491 | -2.7153 | -2.7444 |
|
| 130 |
-
| 0.4211 | 0.77 | 1500 | 0.4645 | -0.6099 | -1.9993 | 0.7656 | 1.3894 | -329.9109 | -299.1959 | -2.6944 | -2.7236 |
|
| 131 |
-
| 0.4931 | 0.83 | 1600 | 0.4684 | -0.6798 | -2.1082 | 0.7656 | 1.4285 | -332.0890 | -300.5934 | -2.7006 | -2.7305 |
|
| 132 |
-
| 0.5029 | 0.88 | 1700 | 0.4595 | -0.5063 | -1.8951 | 0.7812 | 1.3889 | -327.8267 | -297.1233 | -2.7108 | -2.7403 |
|
| 133 |
-
| 0.4965 | 0.93 | 1800 | 0.4613 | -0.5561 | -1.9079 | 0.7812 | 1.3518 | -328.0831 | -298.1203 | -2.7226 | -2.7523 |
|
| 134 |
-
| 0.4337 | 0.98 | 1900 | 0.4608 | -0.5066 | -1.8718 | 0.7656 | 1.3652 | -327.3599 | -297.1296 | -2.7175 | -2.7469 |
|
| 135 |
-
|
| 136 |
-
|
| 137 |
-
### Framework versions
|
| 138 |
-
|
| 139 |
-
- Transformers 4.34.0
|
| 140 |
-
- Pytorch 2.0.1+cu118
|
| 141 |
-
- Datasets 2.12.0
|
| 142 |
-
- Tokenizers 0.14.0
|
|
|
|
| 1 |
+
Created for testing purposes.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|