0430c84d79763b8baf3fd4f79038277b

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

• Loss: 0.8460
• Data Size: 1.0
• Epoch Runtime: 227.5363
• Accuracy: 0.6938
• F1 Macro: 0.6238

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	1.6480	0	12.6451	0.2188	0.1622
No log	1	1973	1.4869	0.0078	15.2358	0.3728	0.1771
0.0312	2	3946	1.0760	0.0156	16.3649	0.5513	0.3373
1.0019	3	5919	0.9153	0.0312	20.6020	0.6123	0.4340
0.936	4	7892	0.8864	0.0625	27.6598	0.6244	0.5582
0.8127	5	9865	0.8390	0.125	40.8108	0.6561	0.5121
0.7991	6	11838	0.7856	0.25	67.1346	0.6764	0.5920
0.8121	7	13811	0.7505	0.5	120.9750	0.6838	0.6112
0.7225	8.0	15784	0.7549	1.0	225.7146	0.6861	0.6217
0.6535	9.0	17757	0.7662	1.0	227.5014	0.7002	0.6131
0.5684	10.0	19730	0.7882	1.0	228.3496	0.6944	0.6285
0.4915	11.0	21703	0.8460	1.0	227.5363	0.6938	0.6238

Framework versions

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