File size: 7,744 Bytes
acf1f27 d2578eb acf1f27 93ec240 acf1f27 93ec240 d2578eb 93ec240 d2578eb 93ec240 acf1f27 93ec240 acf1f27 93ec240 acf1f27 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 |
---
license: llama3
base_model: meta-llama/Meta-Llama-3-8B
tags:
- axolotl
- generated_from_trainer
model-index:
- name: Llama-3-8B-Magpie-Align-SFT-v0.3
results: []
datasets:
- Magpie-Align/Magpie-Reasoning-150K
- Magpie-Align/Magpie-Pro-MT-300K-v0.1
- Magpie-Align/Magpie-Qwen2-Pro-200K-Chinese
language:
- en
- zh
---
# 🐦 Llama-3-8B-Magpie-Align-SFT-v0.3
Project Web: [https://magpie-align.github.io/](https://magpie-align.github.io/)
Arxiv Technical Report: [https://arxiv.org/abs/2406.08464](https://arxiv.org/abs/2406.08464)
Codes: [https://github.com/magpie-align/magpie](https://github.com/magpie-align/magpie)
## 🧐 About This Model
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on
- [Magpie-Align/Magpie-Pro-MT-300K-v0.1](https://huggingface.co/datasets/Magpie-Align/Magpie-Pro-MT-300K-v0.1),
- [Magpie-Align/Magpie-Reasoning-150K](https://huggingface.co/datasets/Magpie-Align/Magpie-Reasoning-150K), and
- [Magpie-Align/Magpie-Qwen2-Pro-200K-Chinese](https://huggingface.co/datasets/Magpie-Align/Magpie-Qwen2-Pro-200K-Chinese)
Compared to [v0.2](https://huggingface.co/Magpie-Align/Llama-3-8B-Magpie-Align-SFT-v0.2), we enhance its multi-lingual ability by incorporating a new dataset with 200K Chinese instructions. It achieves performance comparable with the official Llama-3-8B-Instruct Model **with SFT only**! The detailed benchmark performance is as follows:
- **MT-Bench: 8.050 (1st Turn), 7.350 (Second Turn), 7.700 (Average)**
- **Alpaca Eval 2 (GPT-4-Turbo-1106): 26.37 (LC), 26.42 (WR)**
- **Alpaca Eval 2 (Llama-3-8B-Instruct): 54.53 (LC), 55.26 (WR)**
- **Arena Hard: 20.6**
## 👀 Other Information
**License**: Please follow [Meta Llama 3 Community License](https://llama.meta.com/llama3/license).
**Conversation Template**: Please use Llama 3 **official chat template** for the best performance.
**How to use it?** Please check the official [Llama 3 repository](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct#how-to-use) for detailed instructions. Simply replace the original `model_id` with `Magpie-Align/Llama-3-8B-Magpie-Align-SFT-v0.3`.
## 📚 Citation
If you find the model, data, or code useful, please cite our paper:
```
@article{xu2024magpie,
title={Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing},
author={Zhangchen Xu and Fengqing Jiang and Luyao Niu and Yuntian Deng and Radha Poovendran and Yejin Choi and Bill Yuchen Lin},
year={2024},
eprint={2406.08464},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
**Questions?** Please contact [Zhangchen](https://zhangchenxu.com/) by email.
## Paper Abstract
<details><summary>Click Here</summary>
High-quality instruction data is critical for aligning large language models (LLMs). Although some models, such as Llama-3-Instruct, have open weights, their alignment data remain private, which hinders the democratization of AI. High human labor costs and a limited, predefined scope for prompting prevent existing open-source data creation methods from scaling effectively, potentially limiting the diversity and quality of public alignment datasets. Is it possible to synthesize high-quality instruction data at scale by extracting it directly from an aligned LLM? We present a self-synthesis method for generating large-scale alignment data named Magpie. Our key observation is that aligned LLMs like Llama-3-Instruct can generate a user query when we input only the left-side templates up to the position reserved for user messages, thanks to their auto-regressive nature. We use this method to prompt Llama-3-Instruct and generate 4 million instructions along with their corresponding responses. We perform a comprehensive analysis of the extracted data and select 300K high-quality instances. To compare Magpie data with other public instruction datasets, we fine-tune Llama-3-8B-Base with each dataset and evaluate the performance of the fine-tuned models. Our results indicate that in some tasks, models fine-tuned with Magpie perform comparably to the official Llama-3-8B-Instruct, despite the latter being enhanced with 10 million data points through supervised fine-tuning (SFT) and subsequent feedback learning. We also show that using Magpie solely for SFT can surpass the performance of previous public datasets utilized for both SFT and preference optimization, such as direct preference optimization with UltraFeedback. This advantage is evident on alignment benchmarks such as AlpacaEval, ArenaHard, and WildBench.
</details><be>
## 🏃♂️➡️ Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 32
- total_train_batch_size: 128
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 98
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.8616 | 0.0019 | 1 | 0.8870 |
| 0.5554 | 0.2013 | 106 | 0.5568 |
| 0.5067 | 0.4027 | 212 | 0.5065 |
| 0.4728 | 0.6040 | 318 | 0.4865 |
| 0.4681 | 0.8054 | 424 | 0.4740 |
| 0.4563 | 1.0067 | 530 | 0.4662 |
| 0.4115 | 1.1944 | 636 | 0.4642 |
| 0.3993 | 1.3957 | 742 | 0.4620 |
| 0.4048 | 1.5971 | 848 | 0.4613 |
| 0.4167 | 1.7984 | 954 | 0.4611 |
### Framework versions
- Transformers 4.42.3
- Pytorch 2.3.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
*Internal name for identification: Llama-3-8B-Magpie-Mix-RC*. Please change the model name in the below Axolotl config.
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
base_model: meta-llama/Meta-Llama-3-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: Magpie-Align/Magpie-Reasoning-150K
type: sharegpt
conversation: llama3
- path: Magpie-Align/Magpie-Qwen2-Pro-200K-Chinese
type: sharegpt
conversation: llama3
- path: Magpie-Align/Magpie-Pro-MT-300K-v0.1
type: sharegpt
conversation: llama3
dataset_prepared_path: last_run_prepared
val_set_size: 0.001
output_dir: axolotl_out/Llama-3-8B-Magpie-Mix-RC
sequence_len: 8192
sample_packing: true
eval_sample_packing: false
pad_to_sequence_len: true
wandb_project: SynDa
wandb_entity:
wandb_watch:
wandb_name: Llama-3-8B-Magpie-Mix-RC
wandb_log_model:
hub_model_id: Magpie-Align/Llama-3-8B-Magpie-Mix-RC
gradient_accumulation_steps: 32
micro_batch_size: 1
num_epochs: 2
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 2e-5
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_ratio: 0.1
evals_per_epoch: 5
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
pad_token: <|end_of_text|>
```
</details><br>
|