4361db0dc3f1e2abb8614125784eb8f4

This model is a fine-tuned version of albert/albert-xlarge-v2 on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 1.4830
• Data Size: 1.0
• Epoch Runtime: 561.3408
• Accuracy: 0.3853
• F1 Macro: 0.1112

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	1.5117	0	42.3669	0.3853	0.1118
No log	1	1973	1.5259	0.0078	46.9312	0.3853	0.1112
0.0337	2	3946	1.5430	0.0156	50.4087	0.3853	0.1112
1.5213	3	5919	1.4891	0.0312	58.5779	0.3853	0.1112
1.5089	4	7892	1.4878	0.0625	74.9769	0.3853	0.1112
1.4905	5	9865	1.4932	0.125	107.3085	0.3853	0.1112
1.4903	6	11838	1.4872	0.25	171.7574	0.3853	0.1112
1.4992	7	13811	1.4808	0.5	302.4082	0.3853	0.1112
1.5111	8.0	15784	1.4916	1.0	560.8772	0.3853	0.1112
1.4891	9.0	17757	1.4821	1.0	558.1286	0.3853	0.1112
1.4888	10.0	19730	1.4824	1.0	560.4277	0.3853	0.1112
1.4903	11.0	21703	1.4830	1.0	561.3408	0.3853	0.1112

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.1