feb11d958da949c189638bd4ffaa0f7d

This model is a fine-tuned version of facebook/opt-2.7b on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 1.3009
• Data Size: 1.0
• Epoch Runtime: 901.1093
• Accuracy: 0.6029
• F1 Macro: 0.5342

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	1.9377	0	58.6283	0.1956	0.1373
No log	1	1973	0.9883	0.0078	66.0194	0.5707	0.4695
0.031	2	3946	1.2202	0.0156	76.1170	0.5001	0.3817
1.0719	3	5919	0.9309	0.0312	92.9847	0.6204	0.5012
1.4254	4	7892	1.4194	0.0625	123.1336	0.3979	0.1626
1.1336	5	9865	1.0582	0.125	176.0203	0.5586	0.3675
0.9678	6	11838	0.9375	0.25	275.4740	0.6099	0.4274
0.9487	7	13811	0.8650	0.5	487.8935	0.6419	0.5470
0.8322	8.0	15784	0.8843	1.0	904.0649	0.6317	0.5415
0.7008	9.0	17757	0.9701	1.0	898.8474	0.6263	0.5393
0.5559	10.0	19730	1.0650	1.0	901.7111	0.6012	0.5401
0.4313	11.0	21703	1.3009	1.0	901.1093	0.6029	0.5342

Framework versions

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