Llama-3.2-1B-Summarization-LoRa

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on the scitldr dataset. It achieves the following results on the evaluation set:

• Loss: 2.5661

Model description

Fine-tuned (LoRa) Version of meta-llama/Llama-3.2-1B for Summarization of scientific documents

Intended uses & limitations

Summarization

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Use paged_adamw_32bit 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: 2
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.45	0.2008	200	2.5272
2.4331	0.4016	400	2.5327
2.4369	0.6024	600	2.5285
2.4315	0.8032	800	2.5238
2.4303	1.0040	1000	2.5181
2.1077	1.2048	1200	2.5525
2.0951	1.4056	1400	2.5611
2.0738	1.6064	1600	2.5591
2.0539	1.8072	1800	2.5661

Framework versions

• PEFT 0.13.2
• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3