a58af9b811decb15de671183c4b54b07

This model is a fine-tuned version of albert/albert-base-v1 on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 0.8621
• Data Size: 1.0
• Epoch Runtime: 72.8094
• Accuracy: 0.6829
• F1 Macro: 0.6121

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	1.7944	0	5.8050	0.1102	0.0716
No log	1	1973	1.4652	0.0078	6.7679	0.3875	0.1904
0.0319	2	3946	1.3011	0.0156	7.0154	0.4623	0.2680
1.1769	3	5919	1.1409	0.0312	7.9874	0.5376	0.3510
0.9719	4	7892	0.9420	0.0625	9.9755	0.6041	0.4954
0.8939	5	9865	0.8626	0.125	14.2018	0.6356	0.5305
0.8639	6	11838	0.8189	0.25	22.6114	0.6580	0.5508
0.8509	7	13811	0.7834	0.5	39.4226	0.6698	0.6019
0.7676	8.0	15784	0.7777	1.0	73.1966	0.6729	0.6116
0.6876	9.0	17757	0.7524	1.0	74.0192	0.6952	0.6125
0.6258	10.0	19730	0.7636	1.0	75.1704	0.6802	0.6244
0.6055	11.0	21703	0.7913	1.0	73.2826	0.6909	0.6220
0.5316	12.0	23676	0.7981	1.0	73.7851	0.6908	0.6192
0.4656	13.0	25649	0.8621	1.0	72.8094	0.6829	0.6121

Framework versions

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