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: 2.5290
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: 0.0005
- train_batch_size: 2
- eval_batch_size: 2
- 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 |
---|---|---|---|
2.5409 | 0.3 | 300 | 2.5667 |
2.4932 | 0.6 | 600 | 2.5493 |
2.5285 | 0.9 | 900 | 2.5363 |
2.3828 | 1.2 | 1200 | 2.5399 |
2.402 | 1.51 | 1500 | 2.5358 |
2.3967 | 1.81 | 1800 | 2.5290 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
Model tree for neerajprad/uplimit-project-3-phi-1.5
Base model
microsoft/phi-1_5