aristotle_csv3

This model is a fine-tuned version of ai-forever/rugpt3small_based_on_gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5019

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAFACTOR and the args are: No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.1139	1.0	203	3.7183
3.5598	2.0	406	3.4522
3.2762	3.0	609	3.3321
3.0039	4.0	812	3.3035
2.7239	5.0	1015	3.2801
2.4871	6.0	1218	3.3728
2.277	7.0	1421	3.4348
2.1079	8.0	1624	3.5019

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.0
• Tokenizers 0.21.0