aristotle_new_layer_plain

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: 5.1360

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
8.1202	1.0	203	7.6970
6.5259	2.0	406	6.4359
6.0754	3.0	609	6.0372
5.7242	4.0	812	5.7632
5.2971	5.0	1015	5.5099
5.0427	6.0	1218	5.3732
4.8016	7.0	1421	5.2518
4.559	8.0	1624	5.1812
4.3407	9.0	1827	5.1369
4.0474	10.0	2030	5.1208
3.8746	11.0	2233	5.1177
3.6983	12.0	2436	5.0946
3.5034	13.0	2639	5.1002
3.3277	14.0	2842	5.1041
3.1368	15.0	3045	5.1360

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0