27512031ed97c4bdecb457584e49427b

This model is a fine-tuned version of openai-community/gpt2-medium on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 1.4915
• Data Size: 1.0
• Epoch Runtime: 310.2710
• Accuracy: 0.6894
• F1 Macro: 0.6245

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	9.1634	0	24.2137	0.0951	0.0384
No log	1	1973	1.6523	0.0078	26.3563	0.3821	0.1304
0.0392	2	3946	1.0667	0.0156	28.2697	0.5729	0.3573
0.9577	3	5919	0.8773	0.0312	33.3907	0.6325	0.5078
0.8401	4	7892	0.8125	0.0625	43.5247	0.6524	0.5789
0.8033	5	9865	0.7665	0.125	60.2804	0.6770	0.5769
0.7446	6	11838	0.7603	0.25	96.2873	0.6852	0.5641
0.7378	7	13811	0.7152	0.5	167.2274	0.7024	0.6398
0.6453	8.0	15784	0.7122	1.0	309.2478	0.6991	0.6292
0.5438	9.0	17757	0.7452	1.0	308.7513	0.7039	0.6365
0.3451	10.0	19730	0.9130	1.0	309.3427	0.6820	0.6257
0.281	11.0	21703	1.0554	1.0	310.3317	0.6905	0.6222
0.1528	12.0	23676	1.4915	1.0	310.2710	0.6894	0.6245

Framework versions

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