uplimit-project-3-phi-1.5

This model is a fine-tuned version of microsoft/phi-1_5 on the scitldr dataset. It achieves the following results on the evaluation set:

• Loss: 2.5478

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: 0.001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
2.5968	0.4	800	2.5785
2.5481	0.8	1600	2.5620
2.4473	1.2	2400	2.5622
2.3792	1.61	3200	2.5478

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1