AMANDA

This model is a fine-tuned version of bert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4717
• Accuracy: 0.8243

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: 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: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2389	0.0769	5	0.8651	0.8243
0.6616	0.1538	10	0.6137	0.8243
0.3941	0.2308	15	0.6143	0.8243
0.6576	0.3077	20	0.5270	0.8243
0.4628	0.3846	25	0.4904	0.8243
0.4493	0.4615	30	0.5351	0.8243
0.5603	0.5385	35	0.5049	0.8243
0.5586	0.6154	40	0.4949	0.8243
0.528	0.6923	45	0.4784	0.8243
0.6357	0.7692	50	0.4717	0.8243
0.4228	0.8462	55	0.4674	0.8243
0.4739	0.9231	60	0.4616	0.8243
0.4855	1.0	65	0.4503	0.8243
0.6234	1.0769	70	0.4921	0.8243
0.5158	1.1538	75	0.4351	0.8243
0.3356	1.2308	80	0.4576	0.8243
0.4118	1.3077	85	0.4457	0.8243
0.39	1.3846	90	0.4153	0.8243
0.3848	1.4615	95	0.4377	0.8243
0.3499	1.5385	100	0.4427	0.8209
0.3776	1.6154	105	0.3825	0.8446
0.4228	1.6923	110	0.3755	0.8345
0.3157	1.7692	115	0.4031	0.8243
0.3163	1.8462	120	0.4938	0.8277
0.504	1.9231	125	0.4861	0.8277
0.4722	2.0	130	0.4717	0.8243

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1