Phi-3.5-mini-instruct-qlora
This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on an unknown dataset.
It achieves the following results on the evaluation set:
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 0
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.01
- num_epochs: 4
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
| 1.6254 |
0.3333 |
10 |
1.2928 |
| 1.0852 |
0.6667 |
20 |
0.9771 |
| 0.8786 |
1.0 |
30 |
0.8939 |
| 0.7889 |
1.3333 |
40 |
0.8575 |
| 0.7281 |
1.6667 |
50 |
0.8336 |
| 0.6876 |
2.0 |
60 |
0.8175 |
| 0.6217 |
2.3333 |
70 |
0.8238 |
| 0.6066 |
2.6667 |
80 |
0.8274 |
| 0.614 |
3.0 |
90 |
0.8193 |
| 0.5568 |
3.3333 |
100 |
0.8235 |
| 0.5435 |
3.6667 |
110 |
0.8242 |
| 0.5699 |
4.0 |
120 |
0.8242 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.4.0
- Datasets 3.0.2
- Tokenizers 0.20.0
