PHI30511HMA11H
This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0815
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 100
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.2379 | 0.09 | 10 | 0.5427 |
0.2757 | 0.18 | 20 | 0.1660 |
0.184 | 0.27 | 30 | 0.1553 |
0.1398 | 0.36 | 40 | 0.1268 |
0.1257 | 0.45 | 50 | 0.1158 |
0.1148 | 0.54 | 60 | 0.0949 |
0.0892 | 0.63 | 70 | 0.0841 |
0.0765 | 0.73 | 80 | 0.0660 |
0.0623 | 0.82 | 90 | 0.0698 |
0.0647 | 0.91 | 100 | 0.0660 |
0.0677 | 1.0 | 110 | 0.0672 |
0.0412 | 1.09 | 120 | 0.0798 |
0.0487 | 1.18 | 130 | 0.0708 |
0.0557 | 1.27 | 140 | 0.0685 |
0.0492 | 1.36 | 150 | 0.0652 |
0.05 | 1.45 | 160 | 0.0649 |
0.0484 | 1.54 | 170 | 0.0729 |
0.0468 | 1.63 | 180 | 0.0687 |
0.0433 | 1.72 | 190 | 0.0675 |
0.0484 | 1.81 | 200 | 0.0632 |
0.0433 | 1.9 | 210 | 0.0675 |
0.0452 | 1.99 | 220 | 0.0638 |
0.0216 | 2.08 | 230 | 0.0726 |
0.0164 | 2.18 | 240 | 0.0921 |
0.0159 | 2.27 | 250 | 0.0935 |
0.0122 | 2.36 | 260 | 0.0880 |
0.0215 | 2.45 | 270 | 0.0807 |
0.0134 | 2.54 | 280 | 0.0787 |
0.0115 | 2.63 | 290 | 0.0803 |
0.0171 | 2.72 | 300 | 0.0814 |
0.017 | 2.81 | 310 | 0.0815 |
0.0134 | 2.9 | 320 | 0.0814 |
0.0124 | 2.99 | 330 | 0.0815 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.18.0
- Tokenizers 0.14.1
Model tree for Litzy619/PHI30511HMA11H
Base model
microsoft/Phi-3-mini-4k-instruct