rugpt3large_5epoch_withval

This model is a fine-tuned version of ai-forever/rugpt3large_based_on_gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.6271

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 3
• total_train_batch_size: 24
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.738	1.0000	24244	3.6641
3.7271	2.0000	48488	3.6466
3.7072	3.0	72733	3.6372
3.6851	4.0000	96977	3.6301
3.6827	4.9999	121220	3.6271

Framework versions

• PEFT 0.11.2.dev0
• Transformers 4.42.0.dev0
• Pytorch 2.0.1+cu118
• Tokenizers 0.19.1