calculator_model_test

This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5857

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: 0.001
• train_batch_size: 512
• eval_batch_size: 512
• seed: 42
• 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
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.6713	1.0	6	3.1066
2.6657	2.0	12	2.1494
1.9445	3.0	18	1.7102
1.6766	4.0	24	1.5843
1.5675	5.0	30	1.5400
1.514	6.0	36	1.4790
1.4359	7.0	42	1.3848
1.3846	8.0	48	1.3430
1.3429	9.0	54	1.3113
1.2695	10.0	60	1.2591
1.2346	11.0	66	1.1202
1.1799	12.0	72	1.1988
1.1394	13.0	78	1.0635
1.0823	14.0	84	1.0404
1.0411	15.0	90	1.0058
1.002	16.0	96	0.9632
0.9569	17.0	102	0.9427
0.946	18.0	108	0.8890
0.9083	19.0	114	0.8726
0.8994	20.0	120	0.8342
0.8852	21.0	126	0.8450
0.8647	22.0	132	0.7971
0.8379	23.0	138	0.7805
0.8139	24.0	144	0.7417
0.7895	25.0	150	0.7641
0.8015	26.0	156	0.7093
0.7586	27.0	162	0.7121
0.7653	28.0	168	0.7131
0.7452	29.0	174	0.6806
0.7385	30.0	180	0.6572
0.7004	31.0	186	0.6542
0.6924	32.0	192	0.6507
0.6934	33.0	198	0.6411
0.6803	34.0	204	0.6227
0.6655	35.0	210	0.6102
0.657	36.0	216	0.6125
0.6514	37.0	222	0.5946
0.6528	38.0	228	0.5912
0.6493	39.0	234	0.5893
0.6398	40.0	240	0.5857

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0