ea96bf44b885bcc965b11fa9df05f605

This model is a fine-tuned version of facebook/opt-6.7b on the contemmcm/amazon_reviews_2013 [cell-phone] dataset. It achieves the following results on the evaluation set:

• Loss: 1.7487
• Data Size: 1.0
• Epoch Runtime: 3333.0640
• Accuracy: 0.6055
• F1 Macro: 0.5165

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	2.1502	0	114.1259	0.1655	0.1370
No log	1	1973	1.5965	0.0078	139.3757	0.3851	0.1112
0.0404	2	3946	1.4728	0.0156	175.3895	0.4078	0.2422
1.317	3	5919	1.1682	0.0312	231.5945	0.5148	0.2601
1.0922	4	7892	1.1238	0.0625	331.9087	0.5316	0.4798
0.9894	5	9865	1.0522	0.125	537.8990	0.5864	0.4304
0.9374	6	11838	0.9147	0.25	916.3391	0.6244	0.4894
0.9207	7	13811	0.9228	0.5	1723.0874	0.6134	0.5441
0.8026	8.0	15784	0.8901	1.0	3324.4810	0.6342	0.5411
0.6815	9.0	17757	0.9666	1.0	3324.3755	0.6260	0.5288
0.4927	10.0	19730	1.0969	1.0	3320.1430	0.6149	0.5481
0.338	11.0	21703	1.5839	1.0	3325.9031	0.6123	0.5236
0.2528	12.0	23676	1.7487	1.0	3333.0640	0.6055	0.5165

Framework versions

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