49460af0ac7c5c3c2723b35130ded0e4

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

• Loss: 1.0352
• Data Size: 1.0
• Epoch Runtime: 227.3238
• Accuracy: 0.6782
• F1 Macro: 0.6112

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	1.7093	0	12.6470	0.1858	0.0627
No log	1	1973	1.2258	0.0078	15.0592	0.5262	0.3095
0.0267	2	3946	0.9561	0.0156	16.3004	0.6005	0.4689
0.9322	3	5919	0.9363	0.0312	20.8457	0.6293	0.4825
0.8938	4	7892	0.8451	0.0625	27.7876	0.6460	0.5733
0.8313	5	9865	0.8288	0.125	40.5479	0.6544	0.4942
0.7787	6	11838	0.7628	0.25	67.0522	0.6879	0.5948
0.8367	7	13811	0.7930	0.5	120.6022	0.6861	0.6125
0.7185	8.0	15784	0.7619	1.0	228.1331	0.6891	0.6239
0.6831	9.0	17757	0.7534	1.0	227.5035	0.7005	0.6082
0.5852	10.0	19730	0.8334	1.0	227.2122	0.6534	0.6070
0.582	11.0	21703	0.9040	1.0	228.1371	0.6823	0.5970
0.5017	12.0	23676	0.9457	1.0	226.3781	0.6863	0.6066
0.4286	13.0	25649	1.0352	1.0	227.3238	0.6782	0.6112

Framework versions

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