|
--- |
|
license: mit |
|
base_model: microsoft/Phi-3-mini-4k-instruct |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: PHI30512HMAB18H |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# PHI30512HMAB18H |
|
|
|
This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0740 |
|
|
|
## 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: 60 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 4.2497 | 0.09 | 10 | 0.9067 | |
|
| 0.4537 | 0.18 | 20 | 0.2641 | |
|
| 0.3338 | 0.27 | 30 | 0.2501 | |
|
| 0.2554 | 0.36 | 40 | 0.2343 | |
|
| 0.2601 | 0.45 | 50 | 0.2149 | |
|
| 0.2187 | 0.54 | 60 | 0.2108 | |
|
| 0.2111 | 0.63 | 70 | 0.2032 | |
|
| 0.1904 | 0.73 | 80 | 0.1720 | |
|
| 0.1685 | 0.82 | 90 | 0.1673 | |
|
| 0.1626 | 0.91 | 100 | 0.1636 | |
|
| 0.1578 | 1.0 | 110 | 0.1104 | |
|
| 0.1041 | 1.09 | 120 | 0.0870 | |
|
| 0.0882 | 1.18 | 130 | 0.0854 | |
|
| 0.0933 | 1.27 | 140 | 0.0801 | |
|
| 0.0816 | 1.36 | 150 | 0.0792 | |
|
| 0.0829 | 1.45 | 160 | 0.0751 | |
|
| 0.0741 | 1.54 | 170 | 0.0777 | |
|
| 0.0773 | 1.63 | 180 | 0.0732 | |
|
| 0.0705 | 1.72 | 190 | 0.0705 | |
|
| 0.0749 | 1.81 | 200 | 0.0677 | |
|
| 0.0624 | 1.9 | 210 | 0.0734 | |
|
| 0.0655 | 1.99 | 220 | 0.0722 | |
|
| 0.0436 | 2.08 | 230 | 0.0802 | |
|
| 0.0464 | 2.18 | 240 | 0.0876 | |
|
| 0.0402 | 2.27 | 250 | 0.0758 | |
|
| 0.0379 | 2.36 | 260 | 0.0742 | |
|
| 0.0435 | 2.45 | 270 | 0.0732 | |
|
| 0.0348 | 2.54 | 280 | 0.0766 | |
|
| 0.0356 | 2.63 | 290 | 0.0775 | |
|
| 0.0409 | 2.72 | 300 | 0.0757 | |
|
| 0.0473 | 2.81 | 310 | 0.0746 | |
|
| 0.0401 | 2.9 | 320 | 0.0741 | |
|
| 0.0402 | 2.99 | 330 | 0.0740 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.0 |
|
|
