|
--- |
|
license: mit |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
base_model: microsoft/Phi-3-mini-128k-instruct |
|
model-index: |
|
- name: working |
|
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. --> |
|
|
|
# working |
|
|
|
This model is a fine-tuned version of [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5929 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 6 |
|
- eval_batch_size: 6 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 24 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 2 |
|
- num_epochs: 30 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 3.5322 | 0.86 | 3 | 2.4221 | |
|
| 1.7321 | 2.0 | 7 | 1.6896 | |
|
| 1.8073 | 2.86 | 10 | 1.3296 | |
|
| 0.9839 | 4.0 | 14 | 0.8705 | |
|
| 0.8891 | 4.86 | 17 | 0.6266 | |
|
| 0.4628 | 6.0 | 21 | 0.4525 | |
|
| 0.498 | 6.86 | 24 | 0.4093 | |
|
| 0.3318 | 8.0 | 28 | 0.3812 | |
|
| 0.396 | 8.86 | 31 | 0.3742 | |
|
| 0.2809 | 10.0 | 35 | 0.3603 | |
|
| 0.3487 | 10.86 | 38 | 0.3563 | |
|
| 0.2479 | 12.0 | 42 | 0.3621 | |
|
| 0.3085 | 12.86 | 45 | 0.3734 | |
|
| 0.2225 | 14.0 | 49 | 0.3733 | |
|
| 0.2716 | 14.86 | 52 | 0.3888 | |
|
| 0.1899 | 16.0 | 56 | 0.4287 | |
|
| 0.2319 | 16.86 | 59 | 0.4375 | |
|
| 0.1594 | 18.0 | 63 | 0.4491 | |
|
| 0.1928 | 18.86 | 66 | 0.4811 | |
|
| 0.1307 | 20.0 | 70 | 0.5047 | |
|
| 0.1577 | 20.86 | 73 | 0.5184 | |
|
| 0.1077 | 22.0 | 77 | 0.5539 | |
|
| 0.1333 | 22.86 | 80 | 0.5708 | |
|
| 0.0922 | 24.0 | 84 | 0.5795 | |
|
| 0.1167 | 24.86 | 87 | 0.5875 | |
|
| 0.0818 | 25.71 | 90 | 0.5929 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.10.0 |
|
- Transformers 4.39.3 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |