counterfactual_seed-63_1e-3

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.1845
• Accuracy: 0.4009

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: 0.001
• train_batch_size: 32
• eval_batch_size: 64
• seed: 63
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 32000
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
5.9705	0.9994	1486	4.4079	0.2941
4.3089	1.9992	2972	3.9092	0.3319
3.703	2.9991	4458	3.6332	0.3553
3.5333	3.9996	5945	3.4678	0.3710
3.3129	4.9994	7431	3.3714	0.3801
3.2404	5.9992	8917	3.3118	0.3860
3.1329	6.9991	10403	3.2677	0.3900
3.0918	7.9996	11890	3.2436	0.3927
3.0319	8.9994	13376	3.2273	0.3947
3.0061	9.9992	14862	3.2162	0.3960
2.9698	10.9991	16348	3.2088	0.3971
2.9485	11.9996	17835	3.2029	0.3982
2.9272	12.9994	19321	3.2025	0.3986
2.9081	13.9992	20807	3.1958	0.3994
2.8982	14.9991	22293	3.1934	0.3992
2.8793	15.9996	23780	3.1888	0.3998
2.8779	16.9994	25266	3.1852	0.4005
2.8592	17.9992	26752	3.1844	0.4004
2.8662	18.9991	28238	3.1828	0.4007
2.8451	19.9962	29720	3.1845	0.4009

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.20.0