You can deploy this model and start earning money today!

RichardErkhov/
UCLA-AGI_-_Mistral7B-PairRM-SPPO-Iter1-gguf

Name: UCLA-AGI_-_Mistral7B-PairRM-SPPO-Iter1-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Mistral7B-PairRM-SPPO-Iter1 - GGUF

Model creator: https://huggingface.co/UCLA-AGI/
Original model: https://huggingface.co/UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1/

Name	Quant method	Size
Mistral7B-PairRM-SPPO-Iter1.Q2_K.gguf	Q2_K	2.53GB
Mistral7B-PairRM-SPPO-Iter1.Q3_K_S.gguf	Q3_K_S	2.95GB
Mistral7B-PairRM-SPPO-Iter1.Q3_K.gguf	Q3_K	3.28GB
Mistral7B-PairRM-SPPO-Iter1.Q3_K_M.gguf	Q3_K_M	3.28GB
Mistral7B-PairRM-SPPO-Iter1.Q3_K_L.gguf	Q3_K_L	3.56GB
Mistral7B-PairRM-SPPO-Iter1.IQ4_XS.gguf	IQ4_XS	3.67GB
Mistral7B-PairRM-SPPO-Iter1.Q4_0.gguf	Q4_0	3.83GB
Mistral7B-PairRM-SPPO-Iter1.IQ4_NL.gguf	IQ4_NL	3.87GB
Mistral7B-PairRM-SPPO-Iter1.Q4_K_S.gguf	Q4_K_S	3.86GB
Mistral7B-PairRM-SPPO-Iter1.Q4_K.gguf	Q4_K	4.07GB
Mistral7B-PairRM-SPPO-Iter1.Q4_K_M.gguf	Q4_K_M	4.07GB
Mistral7B-PairRM-SPPO-Iter1.Q4_1.gguf	Q4_1	4.24GB
Mistral7B-PairRM-SPPO-Iter1.Q5_0.gguf	Q5_0	4.65GB
Mistral7B-PairRM-SPPO-Iter1.Q5_K_S.gguf	Q5_K_S	4.65GB
Mistral7B-PairRM-SPPO-Iter1.Q5_K.gguf	Q5_K	4.78GB
Mistral7B-PairRM-SPPO-Iter1.Q5_K_M.gguf	Q5_K_M	4.78GB
Mistral7B-PairRM-SPPO-Iter1.Q5_1.gguf	Q5_1	5.07GB
Mistral7B-PairRM-SPPO-Iter1.Q6_K.gguf	Q6_K	5.53GB
Mistral7B-PairRM-SPPO-Iter1.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0 datasets:

openbmb/UltraFeedback language:
en pipeline_tag: text-generation

Self-Play Preference Optimization for Language Model Alignment (https://arxiv.org/abs/2405.00675)

Mistral7B-PairRM-SPPO-Iter1

This model was developed using Self-Play Preference Optimization at iteration 1, based on the mistralai/Mistral-7B-Instruct-v0.2 architecture as starting point. We utilized the prompt sets from the openbmb/UltraFeedback dataset, splited to 3 parts for 3 iterations by snorkelai/Snorkel-Mistral-PairRM-DPO-Dataset. All responses used are synthetic.

This is the model reported in the paper , with K=5 (generate 5 responses per iteration). We attached the Arena-Hard eval results in this model page.

Links to Other Models

Model Description

Model type: A 7B parameter GPT-like model fine-tuned on synthetic datasets.
Language(s) (NLP): Primarily English
License: Apache-2.0
Finetuned from model: mistralai/Mistral-7B-Instruct-v0.2

AlpacaEval Leaderboard Evaluation Results

Model	LC. Win Rate	Win Rate	Avg. Length
Mistral7B-PairRM-SPPO Iter 1	24.79	23.51	1855
Mistral7B-PairRM-SPPO Iter 2	26.89	27.62	2019
Mistral7B-PairRM-SPPO Iter 3	28.53	31.02	2163
Mistral7B-PairRM-SPPO Iter 1 (best-of-16)	28.71	27.77	1901
Mistral7B-PairRM-SPPO Iter 2 (best-of-16)	31.23	32.12	2035
Mistral7B-PairRM-SPPO Iter 3 (best-of-16)	32.13	34.94	2174

Arena-Hard Evaluation Results

Model	Score	95% CI	average # Tokens
Mistral7B-PairRM-SPPO-Iter3	23.3	(-1.8, 1.8)	578

Open LLM Leaderboard Evaluation Results

Results are reported by using lm-evaluation-harness v0.4.1

	arc_challenge	truthfulqa_mc2	winogrande	gsm8k	hellaswag	mmlu	average
Mistral7B-PairRM-SPPO Iter 1	65.02	69.4	77.82	43.82	85.11	58.84	66.67
Mistral7B-PairRM-SPPO Iter 2	65.53	69.55	77.03	44.35	85.29	58.72	66.75
Mistral7B-PairRM-SPPO Iter 3	65.36	69.97	76.8	42.68	85.16	58.45	66.4

MT-Bench Evaluation Results

	1st Turn	2nd Turn	Average
Mistral7B-PairRM-SPPO Iter 1	7.63	6.79	7.21
Mistral7B-PairRM-SPPO Iter 2	7.90	7.08	7.49
Mistral7B-PairRM-SPPO Iter 3	7.84	7.34	7.59

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-07
eta: 1000
per_device_train_batch_size: 8
gradient_accumulation_steps: 1
seed: 42
distributed_type: deepspeed_zero3
num_devices: 8
optimizer: RMSProp
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_train_epochs: 18.0 (stop at epoch=1.0)

Citation

@misc{wu2024self,
      title={Self-Play Preference Optimization for Language Model Alignment}, 
      author={Wu, Yue and Sun, Zhiqing and Yuan, Huizhuo and Ji, Kaixuan and Yang, Yiming and Gu, Quanquan},
      year={2024},
      eprint={2405.00675},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}

RichardErkhov/UCLA-AGI_-_Mistral7B-PairRM-SPPO-Iter1-gguf