Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


SELM-Llama-3-8B-Instruct-iter-1 - GGUF
- Model creator: https://huggingface.co/ZhangShenao/
- Original model: https://huggingface.co/ZhangShenao/SELM-Llama-3-8B-Instruct-iter-1/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [SELM-Llama-3-8B-Instruct-iter-1.Q2_K.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q2_K.gguf) | Q2_K | 2.96GB |
| [SELM-Llama-3-8B-Instruct-iter-1.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
| [SELM-Llama-3-8B-Instruct-iter-1.IQ3_S.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.IQ3_S.gguf) | IQ3_S | 3.43GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
| [SELM-Llama-3-8B-Instruct-iter-1.IQ3_M.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.IQ3_M.gguf) | IQ3_M | 3.52GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q3_K.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q3_K.gguf) | Q3_K | 3.74GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
| [SELM-Llama-3-8B-Instruct-iter-1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q4_0.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q4_0.gguf) | Q4_0 | 4.34GB |
| [SELM-Llama-3-8B-Instruct-iter-1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q4_K.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q4_K.gguf) | Q4_K | 4.58GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q4_1.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q4_1.gguf) | Q4_1 | 4.78GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q5_0.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q5_0.gguf) | Q5_0 | 5.21GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q5_K.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q5_K.gguf) | Q5_K | 5.34GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q5_1.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q5_1.gguf) | Q5_1 | 5.65GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q6_K.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q6_K.gguf) | Q6_K | 6.14GB |
| [SELM-Llama-3-8B-Instruct-iter-1.Q8_0.gguf](https://huggingface.co/RichardErkhov/ZhangShenao_-_SELM-Llama-3-8B-Instruct-iter-1-gguf/blob/main/SELM-Llama-3-8B-Instruct-iter-1.Q8_0.gguf) | Q8_0 | 7.95GB |




Original model description:
---
license: mit
base_model: meta-llama/Meta-Llama-3-8B-Instruct
tags:
- alignment-handbook
- dpo
- trl
- selm
datasets:
- HuggingFaceH4/ultrafeedback_binarized
model-index:
- name: SELM-Llama-3-8B-Instruct-iter-1
  results: []
---



<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->



[Self-Exploring Language Models: Active Preference Elicitation for Online Alignment](https://arxiv.org/abs/2405.19332).



# SELM-Llama-3-8B-Instruct-iter-1



This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) using synthetic data based on on the HuggingFaceH4/ultrafeedback_binarized dataset.



## Model description



- Model type: A 8B parameter Llama3-instruct-based Self-Exploring Language Models (SELM).
- License: MIT



## Results



|                                        | AlpacaEval 2.0 (LC WR) | MT-Bench (Average) |
|----------------------------------------|------------------------|--------------------|
| [SELM-Llama-3-8B-Instruct-iter-3](https://huggingface.co/ZhangShenao/SELM-Llama-3-8B-Instruct-iter-3)  |    &emsp; &emsp; &emsp;&emsp;           33.47          |   &emsp; &emsp; &emsp;         8.29       |
| [SELM-Llama-3-8B-Instruct-iter-2](https://huggingface.co/ZhangShenao/SELM-Llama-3-8B-Instruct-iter-2) |    &emsp; &emsp; &emsp;&emsp;         35.65         |  &emsp; &emsp; &emsp;         8.09       |
| [SELM-Llama-3-8B-Instruct-iter-1](https://huggingface.co/ZhangShenao/SELM-Llama-3-8B-Instruct-iter-1) |    &emsp; &emsp; &emsp;&emsp;         32.02         |  &emsp; &emsp; &emsp;         7.92       |
| [Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)  |    &emsp; &emsp; &emsp;&emsp;         24.31         |  &emsp; &emsp; &emsp;         7.93       |


### Training hyperparameters

The following hyperparameters were used during training:
- alpha: 0.0001
- beta: 0.01
- train_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- num_epochs: 1

### Framework versions

- Transformers 4.40.2
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.19.1