led_summarizer

This model is a fine-tuned version of allenai/led-base-16384 on the scientific_lay_summarisation dataset. It achieves the following results on the evaluation set:

• Loss: 2.6657

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.2606	0.1067	10	3.0145
3.1127	0.2133	20	2.8998
2.9749	0.32	30	2.8432
2.9502	0.4267	40	2.7848
2.8664	0.5333	50	2.7457
2.8505	0.64	60	2.7244
2.7979	0.7467	70	2.6993
2.778	0.8533	80	2.6769
2.7792	0.96	90	2.6657

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1