gpt2-xl_ft_logits_5k_experiment

This model is a fine-tuned version of gpt2-xl on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 6.8601

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: 64
• total_train_batch_size: 512
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100.0
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.9	7	6.1556
No log	1.9	14	6.3365
No log	2.9	21	6.5909
No log	3.9	28	6.8601

Framework versions

• Transformers 4.17.0
• Pytorch 1.10.0+cu111
• Datasets 2.0.0
• Tokenizers 0.11.6

Perplexity

Score: 17.589759826660156