You can deploy this model and start earning money today!

RichardErkhov/
interview-eval_-_zephyr-7b-stem-case-7-gguf

Name: interview-eval_-_zephyr-7b-stem-case-7-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

zephyr-7b-stem-case-7 - GGUF

Model creator: https://huggingface.co/interview-eval/
Original model: https://huggingface.co/interview-eval/zephyr-7b-stem-case-7/

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

Original model description:

library_name: transformers license: apache-2.0 base_model: alignment-handbook/zephyr-7b-sft-full tags:

alignment-handbook
trl
sft
generated_from_trainer
trl
sft
generated_from_trainer datasets:
EunsuKim/instruct
scottsuk0306/DepthQA model-index:
name: zephyr-7b-stem-case-7 results: []

zephyr-7b-stem-case-7

This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the EunsuKim/instruct and the scottsuk0306/DepthQA datasets. It achieves the following results on the evaluation set:

Loss: 0.0095

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: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 64
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: 0.03
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
1.035	1.0	7	0.8066
0.8612	2.0	14	0.5046
0.4464	3.0	21	0.2836
0.2872	4.0	28	0.1444
0.1132	5.0	35	0.0621
0.0555	6.0	42	0.0280
0.0371	7.0	49	0.0172
0.017	8.0	56	0.0122
0.0126	9.0	63	0.0100
0.0099	10.0	70	0.0095

Framework versions

Transformers 4.44.2
Pytorch 2.4.1+cu121
Datasets 3.0.0
Tokenizers 0.19.1

RichardErkhov/interview-eval_-_zephyr-7b-stem-case-7-gguf

Original model description:

zephyr-7b-stem-case-7

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
interview-eval_-_zephyr-7b-stem-case-7-gguf