jcmei_-_DPO-Zephyr-7B-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
jcmei_-_DPO-Zephyr-7B-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

DPO-Zephyr-7B - GGUF

Model creator: https://huggingface.co/jcmei/
Original model: https://huggingface.co/jcmei/DPO-Zephyr-7B/

Name	Quant method	Size
DPO-Zephyr-7B.Q2_K.gguf	Q2_K	2.53GB
DPO-Zephyr-7B.IQ3_XS.gguf	IQ3_XS	2.81GB
DPO-Zephyr-7B.IQ3_S.gguf	IQ3_S	2.96GB
DPO-Zephyr-7B.Q3_K_S.gguf	Q3_K_S	2.95GB
DPO-Zephyr-7B.IQ3_M.gguf	IQ3_M	3.06GB
DPO-Zephyr-7B.Q3_K.gguf	Q3_K	3.28GB
DPO-Zephyr-7B.Q3_K_M.gguf	Q3_K_M	3.28GB
DPO-Zephyr-7B.Q3_K_L.gguf	Q3_K_L	3.56GB
DPO-Zephyr-7B.IQ4_XS.gguf	IQ4_XS	3.67GB
DPO-Zephyr-7B.Q4_0.gguf	Q4_0	3.83GB
DPO-Zephyr-7B.IQ4_NL.gguf	IQ4_NL	3.87GB
DPO-Zephyr-7B.Q4_K_S.gguf	Q4_K_S	3.86GB
DPO-Zephyr-7B.Q4_K.gguf	Q4_K	4.07GB
DPO-Zephyr-7B.Q4_K_M.gguf	Q4_K_M	4.07GB
DPO-Zephyr-7B.Q4_1.gguf	Q4_1	4.24GB
DPO-Zephyr-7B.Q5_0.gguf	Q5_0	4.65GB
DPO-Zephyr-7B.Q5_K_S.gguf	Q5_K_S	4.65GB
DPO-Zephyr-7B.Q5_K.gguf	Q5_K	4.78GB
DPO-Zephyr-7B.Q5_K_M.gguf	Q5_K_M	4.78GB
DPO-Zephyr-7B.Q5_1.gguf	Q5_1	5.07GB
DPO-Zephyr-7B.Q6_K.gguf	Q6_K	5.53GB
DPO-Zephyr-7B.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: mit base_model: HuggingFaceH4/mistral-7b-sft-beta tags:

alignment-handbook
trl
dpo
generated_from_trainer
trl
dpo
alignment-handbook
generated_from_trainer datasets:
HuggingFaceH4/ultrafeedback_binarized model-index:
name: DPO-Zephyr-7B results: []

DPO-Zephyr-7B

This model is a fine-tuned version of HuggingFaceH4/mistral-7b-sft-beta on the HuggingFaceH4/ultrafeedback_binarized dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-07
train_batch_size: 4
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 256
total_eval_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Framework versions

Transformers 4.40.2
Pytorch 2.1.2+cu121
Datasets 2.14.6
Tokenizers 0.19.1