You can deploy this model and start earning money today!

bailando/
kanana-rag

Name: kanana-rag
Availability: InStock

kakaocorp/kanana-1.5-8b-instruct-2505

apache-2.0

Views

No views yet

kanana-rag

This model is a fine-tuned version of kakaocorp/kanana-1.5-8b-instruct-2505 on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.1031

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-05
train_batch_size: 2
eval_batch_size: 2
seed: 1234
gradient_accumulation_steps: 4
total_train_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 50
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
1.6267	0.3774	5	1.4063
1.6158	0.7547	10	1.4003
1.6861	1.0755	15	1.3906
1.5819	1.4528	20	1.3785
1.6245	1.8302	25	1.3604
1.6763	2.1509	30	1.3386
1.5886	2.5283	35	1.3137
1.4281	2.9057	40	1.2816
1.4478	3.2264	45	1.2471
1.3862	3.6038	50	1.2050
1.4083	3.9811	55	1.1652
1.2984	4.3019	60	1.1351
1.2786	4.6792	65	1.1120
1.2081	5.0	70	1.1031

Framework versions

PEFT 0.18.0
Transformers 4.57.1
Pytorch 2.8.0+cu128
Datasets 4.4.1
Tokenizers 0.22.1

bailando/kanana-rag

kanana-rag

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

bailando/
kanana-rag