This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the belebele_khk_Cyrl_train dataset. It achieves the following results on the evaluation set:

Loss: 0.1161
Accuracy: 0.9716
Mcq Accuracy: 0.5889

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: 0.0002
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 0.1
num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Mcq Accuracy
0.0739	1.0	40	0.0587	0.9747	0.6222
0.0149	2.0	80	0.0605	0.9712	0.6333
0.0083	3.0	120	0.0847	0.9762	0.6222
0.0043	4.0	160	0.1155	0.9702	0.5889
0.0021	5.0	200	0.1161	0.9716	0.5889

Framework versions

PEFT 0.18.1
Transformers 5.2.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

xummer/llama3-1-8b-belebele-lora-khk-cyrl

belebele_khk_Cyrl

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

xummer/
llama3-1-8b-belebele-lora-khk-cyrl