gpt2large-lhm-08

This model was trained from scratch on the training_corpus dataset.

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: 1.8e-05
• train_batch_size: 40
• eval_batch_size: 40
• seed: 42
• gradient_accumulation_steps: 100
• total_train_batch_size: 4000
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1
• Datasets 2.8.0
• Tokenizers 0.13.2