Quantization made by Richard Erkhov.

tulu-2-dpo-13b-ExPO - GGUF

Model creator: https://huggingface.co/chujiezheng/
Original model: https://huggingface.co/chujiezheng/tulu-2-dpo-13b-ExPO/

Name	Quant method	Size
tulu-2-dpo-13b-ExPO.Q2_K.gguf	Q2_K	4.52GB
tulu-2-dpo-13b-ExPO.IQ3_XS.gguf	IQ3_XS	4.99GB
tulu-2-dpo-13b-ExPO.IQ3_S.gguf	IQ3_S	5.27GB
tulu-2-dpo-13b-ExPO.Q3_K_S.gguf	Q3_K_S	5.27GB
tulu-2-dpo-13b-ExPO.IQ3_M.gguf	IQ3_M	5.57GB
tulu-2-dpo-13b-ExPO.Q3_K.gguf	Q3_K	5.9GB
tulu-2-dpo-13b-ExPO.Q3_K_M.gguf	Q3_K_M	5.9GB
tulu-2-dpo-13b-ExPO.Q3_K_L.gguf	Q3_K_L	6.45GB
tulu-2-dpo-13b-ExPO.IQ4_XS.gguf	IQ4_XS	6.54GB
tulu-2-dpo-13b-ExPO.Q4_0.gguf	Q4_0	6.86GB
tulu-2-dpo-13b-ExPO.IQ4_NL.gguf	IQ4_NL	6.9GB
tulu-2-dpo-13b-ExPO.Q4_K_S.gguf	Q4_K_S	6.91GB
tulu-2-dpo-13b-ExPO.Q4_K.gguf	Q4_K	7.33GB
tulu-2-dpo-13b-ExPO.Q4_K_M.gguf	Q4_K_M	7.33GB
tulu-2-dpo-13b-ExPO.Q4_1.gguf	Q4_1	7.61GB
tulu-2-dpo-13b-ExPO.Q5_0.gguf	Q5_0	8.36GB
tulu-2-dpo-13b-ExPO.Q5_K_S.gguf	Q5_K_S	8.36GB
tulu-2-dpo-13b-ExPO.Q5_K.gguf	Q5_K	8.6GB
tulu-2-dpo-13b-ExPO.Q5_K_M.gguf	Q5_K_M	8.6GB
tulu-2-dpo-13b-ExPO.Q5_1.gguf	Q5_1	9.1GB
tulu-2-dpo-13b-ExPO.Q6_K.gguf	Q6_K	9.95GB
tulu-2-dpo-13b-ExPO.Q8_0.gguf	Q8_0	12.88GB

Original model description:

license: other license_name: ai2-impact-license-low-risk license_link: https://allenai.org/impact-license language: - en

tulu-2-dpo-13b-ExPO

The extrapolated (ExPO) model based on allenai/tulu-2-dpo-13b and allenai/tulu-2-13b, as in the "Weak-to-Strong Extrapolation Expedites Alignment" paper.

Specifically, we obtain this model by extrapolating (alpha = 0.5) from the weights of the SFT and DPO/RLHF checkpoints, achieving superior alignment with human preference.

Evaluation Results

Evaluation results on the AlpacaEval 2.0 benchmark (you can find the evaluation outputs on the official GitHub repo):

	Win Rate (Ori)	LC Win Rate (Ori)	Win Rate (+ ExPO)	LC Win Rate (+ ExPO)
`HuggingFaceH4/zephyr-7b-alpha`	6.7%	10.0%	10.6%	13.6%
`HuggingFaceH4/zephyr-7b-beta`	10.2%	13.2%	11.1%	14.0%
`berkeley-nest/Starling-LM-7B-alpha`	15.0%	18.3%	18.2%	19.5%
`Nexusflow/Starling-LM-7B-beta`	26.6%	25.8%	29.6%	26.4%
`snorkelai/Snorkel-Mistral-PairRM`	24.7%	24.0%	28.8%	26.4%
`RLHFlow/LLaMA3-iterative-DPO-final`	29.2%	36.0%	32.7%	37.8%
`internlm/internlm2-chat-1.8b`	3.8%	4.0%	5.2%	4.3%
`internlm/internlm2-chat-7b`	20.5%	18.3%	28.1%	22.7%
`internlm/internlm2-chat-20b`	36.1%	24.9%	46.2%	27.2%
`allenai/tulu-2-dpo-7b`	8.5%	10.2%	11.5%	11.7%
`allenai/tulu-2-dpo-13b`	11.2%	15.5%	15.6%	17.6%
`allenai/tulu-2-dpo-70b`	15.4%	21.2%	23.0%	25.7%

Evaluation results on the MT-Bench benchmark (you can find the evaluation outputs on the official GitHub repo):

	Original	+ ExPO
`HuggingFaceH4/zephyr-7b-alpha`	6.85	6.87
`HuggingFaceH4/zephyr-7b-beta`	7.02	7.06
`berkeley-nest/Starling-LM-7B-alpha`	7.82	7.91
`Nexusflow/Starling-LM-7B-beta`	8.10	8.18
`snorkelai/Snorkel-Mistral-PairRM`	7.63	7.69
`RLHFlow/LLaMA3-iterative-DPO-final`	8.08	8.45
`internlm/internlm2-chat-1.8b`	5.17	5.26
`internlm/internlm2-chat-7b`	7.72	7.80
`internlm/internlm2-chat-20b`	8.13	8.26
`allenai/tulu-2-dpo-7b`	6.35	6.38
`allenai/tulu-2-dpo-13b`	7.00	7.26
`allenai/tulu-2-dpo-70b`	7.79	8.03