dzoQAmodel

This model is a fine-tuned version of sangjeedondrub/tibetan-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.6732

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	16	5.5230
No log	2.0	32	5.3383
No log	3.0	48	5.2939
No log	4.0	64	5.2087
No log	5.0	80	5.1862
No log	6.0	96	5.2205
No log	7.0	112	5.2804
No log	8.0	128	5.3362
No log	9.0	144	5.4572
No log	10.0	160	5.4701
No log	11.0	176	5.5276
No log	12.0	192	5.5678
No log	13.0	208	5.6553
No log	14.0	224	5.6671
No log	15.0	240	5.6732

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1