jungyuko_-_DAVinCI-Yi-Ko-6B-v0.8-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
jungyuko_-_DAVinCI-Yi-Ko-6B-v0.8-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

DAVinCI-Yi-Ko-6B-v0.8 - GGUF

Model creator: https://huggingface.co/jungyuko/
Original model: https://huggingface.co/jungyuko/DAVinCI-Yi-Ko-6B-v0.8/

Name	Quant method	Size
DAVinCI-Yi-Ko-6B-v0.8.Q2_K.gguf	Q2_K	2.24GB
DAVinCI-Yi-Ko-6B-v0.8.IQ3_XS.gguf	IQ3_XS	2.48GB
DAVinCI-Yi-Ko-6B-v0.8.IQ3_S.gguf	IQ3_S	2.6GB
DAVinCI-Yi-Ko-6B-v0.8.Q3_K_S.gguf	Q3_K_S	2.59GB
DAVinCI-Yi-Ko-6B-v0.8.IQ3_M.gguf	IQ3_M	2.69GB
DAVinCI-Yi-Ko-6B-v0.8.Q3_K.gguf	Q3_K	2.86GB
DAVinCI-Yi-Ko-6B-v0.8.Q3_K_M.gguf	Q3_K_M	2.86GB
DAVinCI-Yi-Ko-6B-v0.8.Q3_K_L.gguf	Q3_K_L	3.08GB
DAVinCI-Yi-Ko-6B-v0.8.IQ4_XS.gguf	IQ4_XS	3.18GB
DAVinCI-Yi-Ko-6B-v0.8.Q4_0.gguf	Q4_0	3.32GB
DAVinCI-Yi-Ko-6B-v0.8.IQ4_NL.gguf	IQ4_NL	3.35GB
DAVinCI-Yi-Ko-6B-v0.8.Q4_K_S.gguf	Q4_K_S	3.34GB
DAVinCI-Yi-Ko-6B-v0.8.Q4_K.gguf	Q4_K	3.5GB
DAVinCI-Yi-Ko-6B-v0.8.Q4_K_M.gguf	Q4_K_M	3.5GB
DAVinCI-Yi-Ko-6B-v0.8.Q4_1.gguf	Q4_1	3.66GB
DAVinCI-Yi-Ko-6B-v0.8.Q5_0.gguf	Q5_0	4.0GB
DAVinCI-Yi-Ko-6B-v0.8.Q5_K_S.gguf	Q5_K_S	4.0GB
DAVinCI-Yi-Ko-6B-v0.8.Q5_K.gguf	Q5_K	4.09GB
DAVinCI-Yi-Ko-6B-v0.8.Q5_K_M.gguf	Q5_K_M	4.09GB
DAVinCI-Yi-Ko-6B-v0.8.Q5_1.gguf	Q5_1	4.34GB
DAVinCI-Yi-Ko-6B-v0.8.Q6_K.gguf	Q6_K	4.72GB
DAVinCI-Yi-Ko-6B-v0.8.Q8_0.gguf	Q8_0	6.12GB

Original model description:

license: cc-by-nc-4.0

DAVinCI-Yi-Ko-6B-v0.8

This model is a fine-tuned version of beomi/Yi-Ko-6B on an unknown dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hypuerparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 4
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1.0
mixed_precision_training: Native AMP

Training results

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu121
Datasets 2.0.0
Tokenizers 0.15.0