22b92d1a65652f758ff99ae4de9cc2ee

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

• Loss: 0.9621
• Data Size: 1.0
• Epoch Runtime: 99.6324
• Accuracy: 0.6703
• F1 Macro: 0.6084

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	1.7292	0	8.5793	0.1016	0.0757
No log	1	1973	1.4075	0.0078	9.7491	0.4372	0.2872
0.0283	2	3946	1.0022	0.0156	9.9230	0.5882	0.4722
0.9757	3	5919	0.9933	0.0312	11.3562	0.5797	0.4497
0.9141	4	7892	0.8852	0.0625	14.2516	0.6413	0.5391
0.8647	5	9865	0.8909	0.125	19.9241	0.6265	0.4779
0.8128	6	11838	0.8286	0.25	31.2026	0.6516	0.5594
0.8844	7	13811	0.8028	0.5	53.7901	0.6678	0.5719
0.8186	8.0	15784	0.8681	1.0	98.8104	0.6464	0.5306
0.6741	9.0	17757	0.8206	1.0	99.7908	0.6727	0.5613
0.7092	10.0	19730	0.7418	1.0	98.8759	0.6862	0.6307
0.5733	11.0	21703	0.8961	1.0	100.0342	0.6688	0.5870
0.4992	12.0	23676	0.8265	1.0	99.1054	0.6868	0.6042
0.41	13.0	25649	0.8860	1.0	99.3899	0.6824	0.6128
0.3838	14.0	27622	0.9621	1.0	99.6324	0.6703	0.6084

Framework versions

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