continue_pretrain_t5_base_10tokens

This model is a fine-tuned version of google-t5/t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 5.0015
• Rouge: {'rouge1': 0.15180664921665177, 'rouge2': 0.14241763000023774, 'rougeL': 0.1513952140128575, 'rougeLsum': 0.1517189553463021}
• Exact Match: {'exact_match': 0.0}

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: 2e-05
• train_batch_size: 14
• eval_batch_size: 14
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 28
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge	Exact Match
0.1039	1.0	1786	4.8370	{'rouge1': 0.08468086761955296, 'rouge2': 0.07928600729695852, 'rougeL': 0.08453698334268148, 'rougeLsum': 0.08473098923719942}	{'exact_match': 0.0}
0.0503	2.0	3572	4.9960	{'rouge1': 0.15080775818986453, 'rouge2': 0.1414201900527639, 'rougeL': 0.15034957154685738, 'rougeLsum': 0.1506605398259596}	{'exact_match': 0.0}
0.0521	3.0	5358	5.0015	{'rouge1': 0.15180664921665177, 'rouge2': 0.14241763000023774, 'rougeL': 0.1513952140128575, 'rougeLsum': 0.1517189553463021}	{'exact_match': 0.0}

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.2+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1