791bd783bb7d470da6c511984e8e4be4

This model is a fine-tuned version of google/gemma-2b on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 4.4201
• Data Size: 1.0
• Epoch Runtime: 748.2557
• Accuracy: 0.6716
• F1 Macro: 0.5990

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	14.6129	0	45.1468	0.1232	0.0703
No log	1	1973	4.1900	0.0078	50.6193	0.5434	0.4874
0.113	2	3946	3.8750	0.0156	57.5991	0.5953	0.4168
4.0048	3	5919	3.3635	0.0312	72.0298	0.6585	0.5030
3.4985	4	7892	3.2877	0.0625	96.6778	0.6466	0.5601
3.3088	5	9865	3.2563	0.125	139.4303	0.6576	0.5131
3.0347	6	11838	3.2212	0.25	224.8132	0.6696	0.5134
3.1381	7	13811	3.1242	0.5	401.7742	0.6743	0.6135
2.8412	8.0	15784	2.9814	1.0	752.0521	0.6865	0.6182
2.4671	9.0	17757	3.0518	1.0	747.1279	0.6945	0.6074
1.957	10.0	19730	3.1940	1.0	748.5617	0.6881	0.6228
1.4056	11.0	21703	4.4067	1.0	748.1682	0.6693	0.5912
0.9514	12.0	23676	4.4201	1.0	748.2557	0.6716	0.5990

Framework versions

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