dnd2

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

• Loss: 0.1599
• Accuracy: 0.9298

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-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	15	0.9829	0.6316
No log	2.0	30	0.5511	0.7193
No log	3.0	45	0.3689	0.8421
No log	4.0	60	0.2890	0.9123
No log	5.0	75	0.2439	0.8947
No log	6.0	90	0.2130	0.8947
No log	7.0	105	0.1932	0.8947
No log	8.0	120	0.1789	0.9123
No log	9.0	135	0.1702	0.9123
No log	10.0	150	0.1643	0.9123
No log	11.0	165	0.1610	0.9123
No log	12.0	180	0.1599	0.9298

Framework versions

• Transformers 4.32.1
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.13.2