|
--- |
|
license: mit |
|
base_model: microsoft/Phi-3-mini-4k-instruct |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: PHI30512HMAB7H |
|
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. --> |
|
|
|
# PHI30512HMAB7H |
|
|
|
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: 0.1660 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 4.9249 | 0.09 | 10 | 2.4664 | |
|
| 1.3773 | 0.18 | 20 | 0.4052 | |
|
| 0.394 | 0.27 | 30 | 0.3430 | |
|
| 2.5727 | 0.36 | 40 | 0.3864 | |
|
| 0.2708 | 0.45 | 50 | 0.1622 | |
|
| 0.1648 | 0.54 | 60 | 0.1491 | |
|
| 0.1254 | 0.63 | 70 | 0.1389 | |
|
| 0.1202 | 0.73 | 80 | 0.1068 | |
|
| 0.092 | 0.82 | 90 | 0.0929 | |
|
| 0.097 | 0.91 | 100 | 0.0817 | |
|
| 0.0815 | 1.0 | 110 | 0.0795 | |
|
| 0.0971 | 1.09 | 120 | 0.1372 | |
|
| 0.3829 | 1.18 | 130 | 0.2014 | |
|
| 0.2626 | 1.27 | 140 | 0.1422 | |
|
| 0.1206 | 1.36 | 150 | 0.1053 | |
|
| 2.8589 | 1.45 | 160 | 2.3060 | |
|
| 1.7749 | 1.54 | 170 | 1.1543 | |
|
| 0.8021 | 1.63 | 180 | 0.5702 | |
|
| 0.464 | 1.72 | 190 | 0.3593 | |
|
| 0.3491 | 1.81 | 200 | 0.3201 | |
|
| 0.3161 | 1.9 | 210 | 0.3053 | |
|
| 0.2851 | 1.99 | 220 | 0.2623 | |
|
| 0.2537 | 2.08 | 230 | 0.2722 | |
|
| 0.244 | 2.18 | 240 | 0.1909 | |
|
| 0.1926 | 2.27 | 250 | 0.1829 | |
|
| 0.1805 | 2.36 | 260 | 0.1712 | |
|
| 0.1712 | 2.45 | 270 | 0.1778 | |
|
| 0.1665 | 2.54 | 280 | 0.1669 | |
|
| 0.1733 | 2.63 | 290 | 0.1705 | |
|
| 0.173 | 2.72 | 300 | 0.1668 | |
|
| 0.1687 | 2.81 | 310 | 0.1676 | |
|
| 0.1673 | 2.9 | 320 | 0.1662 | |
|
| 0.1684 | 2.99 | 330 | 0.1660 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.0 |
|
|
