e02eb2c284892973e8f25a68015d5374

This model is a fine-tuned version of google-bert/bert-large-uncased-whole-word-masking on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 0.8726
• Data Size: 1.0
• Epoch Runtime: 224.2795
• Accuracy: 0.6757
• F1 Macro: 0.5942

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	1.6530	0	12.5811	0.1864	0.0929
No log	1	1973	1.3305	0.0078	15.4530	0.4914	0.3185
0.0282	2	3946	0.9280	0.0156	16.3489	0.6041	0.4319
0.9532	3	5919	0.8946	0.0312	20.7930	0.6256	0.4793
0.8683	4	7892	0.8571	0.0625	27.4441	0.6392	0.5369
0.8159	5	9865	0.8939	0.125	40.5887	0.6445	0.4911
0.7847	6	11838	0.7965	0.25	66.6236	0.6682	0.5626
0.819	7	13811	0.7356	0.5	119.7266	0.6917	0.6191
0.7425	8.0	15784	0.7798	1.0	224.4164	0.6764	0.6281
0.6695	9.0	17757	0.8099	1.0	223.5268	0.6943	0.6024
0.5917	10.0	19730	0.7884	1.0	223.4544	0.6763	0.6180
0.5931	11.0	21703	0.8726	1.0	224.2795	0.6757	0.5942

Framework versions

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