ale-bay_-_zephyr-2b-gemma-dft-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
ale-bay_-_zephyr-2b-gemma-dft-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

zephyr-2b-gemma-dft - GGUF

Model creator: https://huggingface.co/ale-bay/
Original model: https://huggingface.co/ale-bay/zephyr-2b-gemma-dft/

Name	Quant method	Size
zephyr-2b-gemma-dft.Q2_K.gguf	Q2_K	1.08GB
zephyr-2b-gemma-dft.Q3_K_S.gguf	Q3_K_S	1.2GB
zephyr-2b-gemma-dft.Q3_K.gguf	Q3_K	1.29GB
zephyr-2b-gemma-dft.Q3_K_M.gguf	Q3_K_M	1.29GB
zephyr-2b-gemma-dft.Q3_K_L.gguf	Q3_K_L	1.36GB
zephyr-2b-gemma-dft.IQ4_XS.gguf	IQ4_XS	1.4GB
zephyr-2b-gemma-dft.Q4_0.gguf	Q4_0	1.44GB
zephyr-2b-gemma-dft.IQ4_NL.gguf	IQ4_NL	1.45GB
zephyr-2b-gemma-dft.Q4_K_S.gguf	Q4_K_S	1.45GB
zephyr-2b-gemma-dft.Q4_K.gguf	Q4_K	1.52GB
zephyr-2b-gemma-dft.Q4_K_M.gguf	Q4_K_M	1.52GB
zephyr-2b-gemma-dft.Q4_1.gguf	Q4_1	1.56GB
zephyr-2b-gemma-dft.Q5_0.gguf	Q5_0	1.68GB
zephyr-2b-gemma-dft.Q5_K_S.gguf	Q5_K_S	1.68GB
zephyr-2b-gemma-dft.Q5_K.gguf	Q5_K	1.71GB
zephyr-2b-gemma-dft.Q5_K_M.gguf	Q5_K_M	1.71GB
zephyr-2b-gemma-dft.Q5_1.gguf	Q5_1	1.79GB
zephyr-2b-gemma-dft.Q6_K.gguf	Q6_K	1.92GB
zephyr-2b-gemma-dft.Q8_0.gguf	Q8_0	2.49GB

Original model description:

base_model: data/gemma-2b tags:

alignment-handbook
trl
sft
generated_from_trainer
trl
sft
generated_from_trainer datasets:
argilla/dpo-mix-7k model-index:
name: zephyr-2b-gemma-dft results: []

Visualize in Weights & Biases

zephyr-2b-gemma-dft

This model is a fine-tuned version of data/gemma-2b on the argilla/dpo-mix-7k dataset. It achieves the following results on the evaluation set:

Loss: 0.0001

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: 2e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 0

Training results

Framework versions

Transformers 4.42.4
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1