electra-large-discriminator_LOGIC_Native

This model is a fine-tuned version of google/electra-large-discriminator on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.0208
• Accuracy: 0.6833
• Macro Precision: 0.6729
• Macro F1: 0.6301

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro Precision	Macro F1
No log	1.0	116	2.4019	0.29	0.2171	0.1836
No log	2.0	232	1.8764	0.4767	0.4286	0.4022
No log	3.0	348	1.5123	0.6167	0.5440	0.5488
No log	4.0	464	1.4121	0.6367	0.5851	0.5757
1.7300	5.0	580	1.3948	0.6333	0.6169	0.5934
1.7300	6.0	696	1.4507	0.6833	0.6637	0.6408
1.7300	7.0	812	1.6601	0.65	0.6276	0.6079
1.7300	8.0	928	1.7656	0.6667	0.6832	0.6241
0.3003	9.0	1044	1.8199	0.67	0.6249	0.6087
0.3003	10.0	1160	1.9297	0.67	0.6601	0.6248
0.3003	11.0	1276	1.9762	0.68	0.6649	0.6310
0.3003	12.0	1392	2.0208	0.6833	0.6729	0.6301

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2