stance_baseline_v3

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

• Loss: 1.7193
• Accuracy Stance: 0.6689
• F1 Macro Stance: 0.4543
• F1 Weighted Stance: 0.6476
• F1 Macro Total: 0.4543
• F1 Weighted Total: 0.6476

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: 6e-06
• train_batch_size: 4
• eval_batch_size: 4
• seed: 1337
• 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
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Stance	F1 Macro Stance	F1 Weighted Stance	F1 Macro Total	F1 Weighted Total
No log	1.0	40	1.0642	0.4189	0.2040	0.2589	0.2040	0.2589
No log	2.0	80	1.0381	0.4189	0.1968	0.2474	0.1968	0.2474
No log	3.0	120	1.0219	0.4189	0.1968	0.2474	0.1968	0.2474
No log	4.0	160	1.0082	0.4932	0.3289	0.4549	0.3289	0.4549
No log	5.0	200	0.9815	0.5473	0.3761	0.5293	0.3761	0.5293
No log	6.0	240	0.9725	0.5743	0.3877	0.5531	0.3877	0.5531
No log	7.0	280	0.9607	0.6216	0.3949	0.5701	0.3949	0.5701
No log	8.0	320	0.9654	0.5946	0.4179	0.5941	0.4179	0.5941
No log	9.0	360	1.0517	0.6216	0.3984	0.5760	0.3984	0.5760
No log	10.0	400	1.0698	0.6284	0.4185	0.6026	0.4185	0.6026
No log	11.0	440	1.1862	0.6554	0.4445	0.6340	0.4445	0.6340
No log	12.0	480	1.3383	0.6419	0.4357	0.6237	0.4357	0.6237
0.7445	13.0	520	1.5536	0.6216	0.4219	0.6064	0.4219	0.6064
0.7445	14.0	560	1.6323	0.6554	0.4444	0.6347	0.4444	0.6347
0.7445	15.0	600	1.7193	0.6689	0.4543	0.6476	0.4543	0.6476
0.7445	16.0	640	1.8830	0.6486	0.4372	0.6248	0.4372	0.6248
0.7445	17.0	680	1.8240	0.6149	0.4266	0.6111	0.4266	0.6111
0.7445	18.0	720	2.0342	0.6419	0.4338	0.6217	0.4338	0.6217
0.7445	19.0	760	1.9920	0.6351	0.4379	0.6263	0.4379	0.6263
0.7445	20.0	800	2.0674	0.6486	0.4430	0.6338	0.4430	0.6338

Framework versions

• Transformers 4.50.2
• Pytorch 2.6.0+cu124
• Datasets 3.0.1
• Tokenizers 0.21.1