gpt2_512_nw

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7899

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: 0.0005
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.2012	0.96	500	2.7336
2.5917	1.92	1000	2.1248
2.2084	2.88	1500	1.8910
1.998	3.84	2000	1.7899

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0