19baa7b9dae90eabb07224a6ac31098b

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

• Loss: 1.2118
• Data Size: 1.0
• Epoch Runtime: 69.7131
• Accuracy: 0.6550
• F1 Macro: 0.5886

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	1.5968	0	5.2112	0.2384	0.1117
No log	1	1973	1.5006	0.0078	6.0295	0.3899	0.1387
0.0316	2	3946	1.3779	0.0156	6.1610	0.4298	0.2083
1.2476	3	5919	1.1972	0.0312	7.1579	0.4987	0.2796
1.0789	4	7892	1.0345	0.0625	9.2532	0.5612	0.4372
0.9288	5	9865	0.8951	0.125	13.3585	0.6249	0.5314
0.8727	6	11838	0.8392	0.25	21.3606	0.6523	0.5580
0.8609	7	13811	0.8048	0.5	38.3610	0.6623	0.5926
0.7663	8.0	15784	0.8070	1.0	69.7185	0.6554	0.6037
0.6575	9.0	17757	0.8013	1.0	68.6924	0.6766	0.5995
0.5312	10.0	19730	0.9003	1.0	69.6074	0.6427	0.5930
0.4557	11.0	21703	0.9741	1.0	70.2688	0.6566	0.5940
0.3856	12.0	23676	1.0758	1.0	69.6125	0.6596	0.5898
0.3078	13.0	25649	1.2118	1.0	69.7131	0.6550	0.5886

Framework versions

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