You can deploy this model and start earning money today!

RichardErkhov/
tongliuphysics_-_zephyr0-7b-ultra-p-0.02-gguf

Name: tongliuphysics_-_zephyr0-7b-ultra-p-0.02-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

zephyr0-7b-ultra-p-0.02 - GGUF

Model creator: https://huggingface.co/tongliuphysics/
Original model: https://huggingface.co/tongliuphysics/zephyr0-7b-ultra-p-0.02/

Name	Quant method	Size
zephyr0-7b-ultra-p-0.02.Q2_K.gguf	Q2_K	2.53GB
zephyr0-7b-ultra-p-0.02.IQ3_XS.gguf	IQ3_XS	2.81GB
zephyr0-7b-ultra-p-0.02.IQ3_S.gguf	IQ3_S	2.96GB
zephyr0-7b-ultra-p-0.02.Q3_K_S.gguf	Q3_K_S	2.95GB
zephyr0-7b-ultra-p-0.02.IQ3_M.gguf	IQ3_M	3.06GB
zephyr0-7b-ultra-p-0.02.Q3_K.gguf	Q3_K	3.28GB
zephyr0-7b-ultra-p-0.02.Q3_K_M.gguf	Q3_K_M	3.28GB
zephyr0-7b-ultra-p-0.02.Q3_K_L.gguf	Q3_K_L	3.56GB
zephyr0-7b-ultra-p-0.02.IQ4_XS.gguf	IQ4_XS	3.67GB
zephyr0-7b-ultra-p-0.02.Q4_0.gguf	Q4_0	3.83GB
zephyr0-7b-ultra-p-0.02.IQ4_NL.gguf	IQ4_NL	3.87GB
zephyr0-7b-ultra-p-0.02.Q4_K_S.gguf	Q4_K_S	3.86GB
zephyr0-7b-ultra-p-0.02.Q4_K.gguf	Q4_K	4.07GB
zephyr0-7b-ultra-p-0.02.Q4_K_M.gguf	Q4_K_M	4.07GB
zephyr0-7b-ultra-p-0.02.Q4_1.gguf	Q4_1	4.24GB
zephyr0-7b-ultra-p-0.02.Q5_0.gguf	Q5_0	4.65GB
zephyr0-7b-ultra-p-0.02.Q5_K_S.gguf	Q5_K_S	4.65GB
zephyr0-7b-ultra-p-0.02.Q5_K.gguf	Q5_K	4.78GB
zephyr0-7b-ultra-p-0.02.Q5_K_M.gguf	Q5_K_M	4.78GB
zephyr0-7b-ultra-p-0.02.Q5_1.gguf	Q5_1	5.07GB
zephyr0-7b-ultra-p-0.02.Q6_K.gguf	Q6_K	5.53GB
zephyr0-7b-ultra-p-0.02.Q8_0.gguf	Q8_0	7.17GB

Original model description:

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

trl
dpo
generated_from_trainer model-index:
name: zephyr0-7b-ultra-p-0.02 results: []

zephyr0-7b-ultra-p-0.02

This model is a fine-tuned version of HuggingFaceH4/mistral-7b-sft-beta on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5172
Rewards/chosen: -1.0774
Rewards/rejected: -1.8857
Rewards/accuracies: 0.75
Rewards/margins: 0.8083
Logps/rejected: -416.3554
Logps/chosen: -330.1051
Logits/rejected: -1.9304
Logits/chosen: -2.0243

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: 2
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 128
total_eval_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.576	0.2060	100	0.5490	-0.7973	-1.3203	0.7266	0.5230	-359.8105	-302.0933	-2.4928	-2.5572
0.5325	0.4119	200	0.5335	-1.0228	-1.6514	0.7266	0.6287	-392.9271	-324.6441	-2.1888	-2.2654
0.5083	0.6179	300	0.5271	-0.9857	-1.7625	0.7422	0.7768	-404.0363	-320.9385	-2.0319	-2.1170
0.5073	0.8239	400	0.5199	-1.0283	-1.8048	0.75	0.7765	-408.2682	-325.1979	-1.9835	-2.0746

Framework versions

Transformers 4.45.1
Pytorch 2.4.1+cu121
Datasets 3.0.0
Tokenizers 0.20.0

RichardErkhov/tongliuphysics_-_zephyr0-7b-ultra-p-0.02-gguf

Original model description:

zephyr0-7b-ultra-p-0.02

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
tongliuphysics_-_zephyr0-7b-ultra-p-0.02-gguf