|
Quantization made by Richard Erkhov. |
|
|
|
[Github](https://github.com/RichardErkhov) |
|
|
|
[Discord](https://discord.gg/pvy7H8DZMG) |
|
|
|
[Request more models](https://github.com/RichardErkhov/quant_request) |
|
|
|
|
|
zephyr-7b-sft-full-SPIN-iter0 - GGUF |
|
- Model creator: https://huggingface.co/UCLA-AGI/ |
|
- Original model: https://huggingface.co/UCLA-AGI/zephyr-7b-sft-full-SPIN-iter0/ |
|
|
|
|
|
| Name | Quant method | Size | |
|
| ---- | ---- | ---- | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q2_K.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q2_K.gguf) | Q2_K | 2.53GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.IQ3_XS.gguf) | IQ3_XS | 2.81GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.IQ3_S.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.IQ3_S.gguf) | IQ3_S | 2.96GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q3_K_S.gguf) | Q3_K_S | 2.95GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.IQ3_M.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.IQ3_M.gguf) | IQ3_M | 3.06GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q3_K.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q3_K.gguf) | Q3_K | 3.28GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q3_K_M.gguf) | Q3_K_M | 3.28GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q3_K_L.gguf) | Q3_K_L | 3.56GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.IQ4_XS.gguf) | IQ4_XS | 3.67GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q4_0.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q4_0.gguf) | Q4_0 | 3.83GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.IQ4_NL.gguf) | IQ4_NL | 3.87GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q4_K_S.gguf) | Q4_K_S | 3.86GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q4_K.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q4_K.gguf) | Q4_K | 4.07GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q4_K_M.gguf) | Q4_K_M | 4.07GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q4_1.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q4_1.gguf) | Q4_1 | 4.24GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q5_0.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q5_0.gguf) | Q5_0 | 4.65GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q5_K_S.gguf) | Q5_K_S | 4.65GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q5_K.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q5_K.gguf) | Q5_K | 4.78GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q5_K_M.gguf) | Q5_K_M | 4.78GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q5_1.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q5_1.gguf) | Q5_1 | 5.07GB | |
|
| [zephyr-7b-sft-full-SPIN-iter0.Q6_K.gguf](https://huggingface.co/RichardErkhov/UCLA-AGI_-_zephyr-7b-sft-full-SPIN-iter0-gguf/blob/main/zephyr-7b-sft-full-SPIN-iter0.Q6_K.gguf) | Q6_K | 5.53GB | |
|
|
|
|
|
|
|
|
|
Original model description: |
|
--- |
|
license: mit |
|
datasets: |
|
- UCLA-AGI/SPIN_iter0 |
|
language: |
|
- en |
|
pipeline_tag: text-generation |
|
--- |
|
Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models (https://arxiv.org/abs/2401.01335) |
|
|
|
# zephyr-7b-sft-full-spin-iter0 |
|
|
|
This model is a self-play fine-tuned model at iteration 0 from [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) using synthetic data based on on the [HuggingFaceH4/ultrachat_200k](https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k) dataset. |
|
|
|
## Model Details |
|
|
|
### Model Description |
|
|
|
- Model type: A 7B parameter GPT-like model fine-tuned on synthetic datasets. |
|
- Language(s) (NLP): Primarily English |
|
- License: MIT |
|
- Finetuned from model: alignment-handbook/zephyr-7b-sft-full (based on mistralai/Mistral-7B-v0.1) |
|
|
|
### Training hyperparameters |
|
The following hyperparameters were used during training: |
|
|
|
- learning_rate: 5e-07 |
|
- train_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 8 |
|
- total_train_batch_size: 64 |
|
- optimizer: RMSProp |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 2.0 |
|
|
|
## [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) |
|
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_UCLA-AGI__test0) |
|
| Metric | Value | |
|
|-----------------------|---------------------------| |
|
| Avg. | 62.37 | |
|
| ARC (25-shot) | 63.65 | |
|
| HellaSwag (10-shot) | 84.44 | |
|
| MMLU (5-shot) | 61.01 | |
|
| TruthfulQA (0-shot) | 50.48 | |
|
| Winogrande (5-shot) | 77.98 | |
|
| GSM8K (5-shot) | 36.69 | |
|
|
|
## Citation |
|
``` |
|
@misc{chen2024selfplay, |
|
title={Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models}, |
|
author={Zixiang Chen and Yihe Deng and Huizhuo Yuan and Kaixuan Ji and Quanquan Gu}, |
|
year={2024}, |
|
eprint={2401.01335}, |
|
archivePrefix={arXiv}, |
|
primaryClass={cs.LG} |
|
} |
|
``` |
|
|
|
|