metadata
license: other
base_model: microsoft/phi-1_5
tags:
- generated_from_trainer
datasets:
- scitldr
model-index:
- name: arphiv
results: []
arphiv
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: 1.9806
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 |
|---|---|---|---|
| 1.9802 | 0.1 | 200 | 2.0909 |
| 1.9764 | 0.2 | 400 | 2.0620 |
| 1.9247 | 0.3 | 600 | 2.0534 |
| 2.0485 | 0.4 | 800 | 2.0619 |
| 1.9351 | 0.5 | 1000 | 2.0143 |
| 1.9793 | 0.6 | 1200 | 2.0255 |
| 1.9704 | 0.7 | 1400 | 2.0190 |
| 1.9144 | 0.8 | 1600 | 1.9922 |
| 2.0139 | 0.9 | 1800 | 2.0024 |
| 1.9789 | 1.0 | 2000 | 1.9845 |
| 1.622 | 1.1 | 2200 | 2.0068 |
| 1.6057 | 1.2 | 2400 | 2.0123 |
| 1.6747 | 1.31 | 2600 | 1.9998 |
| 1.5369 | 1.41 | 2800 | 2.0055 |
| 1.5566 | 1.51 | 3000 | 2.0138 |
| 1.3799 | 1.61 | 3200 | 1.9924 |
| 1.5956 | 1.71 | 3400 | 1.9849 |
| 1.3932 | 1.81 | 3600 | 1.9926 |
| 1.4757 | 1.91 | 3800 | 1.9806 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1