deberta-base-azerbaijani-v2

This model is a fine-tuned version of microsoft/mdeberta-v3-base on a various cleaned community corpus. It achieves the following results on the evaluation set:

• Loss: 0.9572

We thank Microsoft Accelerating Foundation Models Research Program for supporting our research. Authors: Mammad Hajili, Duygu Ataman

Model description

The model was trained on masked language model task on a 4 X A100 80GB GPU for 23 hours. For downstream tasks, it requires to be fine-tuned based on objective of the task.

Training and evaluation data

The training data is clean mix of various Azerbaijani corpus shared by the community.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 128
• total_eval_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Perplexity at epoch 5: 2.6

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1