|
--- |
|
license: mit |
|
base_model: microsoft/Phi-3-mini-4k-instruct |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: PHI30512HMAB22H |
|
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. --> |
|
|
|
# PHI30512HMAB22H |
|
|
|
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: 0.0375 |
|
|
|
## 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: 80 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 4.9533 | 0.09 | 10 | 1.5870 | |
|
| 0.7577 | 0.18 | 20 | 0.2615 | |
|
| 0.2902 | 0.27 | 30 | 0.2473 | |
|
| 0.2709 | 0.36 | 40 | 0.2446 | |
|
| 0.26 | 0.45 | 50 | 0.2296 | |
|
| 0.2324 | 0.54 | 60 | 0.2222 | |
|
| 0.2278 | 0.63 | 70 | 0.2435 | |
|
| 0.236 | 0.73 | 80 | 0.2284 | |
|
| 0.1862 | 0.82 | 90 | 0.1793 | |
|
| 0.1768 | 0.91 | 100 | 0.1645 | |
|
| 0.1688 | 1.0 | 110 | 0.1509 | |
|
| 0.1331 | 1.09 | 120 | 0.1000 | |
|
| 0.0923 | 1.18 | 130 | 0.1010 | |
|
| 0.097 | 1.27 | 140 | 0.0757 | |
|
| 0.0803 | 1.36 | 150 | 0.0747 | |
|
| 0.0813 | 1.45 | 160 | 0.0709 | |
|
| 0.0747 | 1.54 | 170 | 0.0715 | |
|
| 0.0726 | 1.63 | 180 | 0.0665 | |
|
| 0.0678 | 1.72 | 190 | 0.0680 | |
|
| 0.0692 | 1.81 | 200 | 0.0700 | |
|
| 0.0607 | 1.9 | 210 | 0.0704 | |
|
| 0.064 | 1.99 | 220 | 0.0667 | |
|
| 0.0417 | 2.08 | 230 | 0.0758 | |
|
| 0.0415 | 2.18 | 240 | 0.0743 | |
|
| 0.0356 | 2.27 | 250 | 0.0644 | |
|
| 0.0321 | 2.36 | 260 | 0.0600 | |
|
| 0.0365 | 2.45 | 270 | 0.0490 | |
|
| 0.0255 | 2.54 | 280 | 0.0453 | |
|
| 0.0239 | 2.63 | 290 | 0.0437 | |
|
| 0.0299 | 2.72 | 300 | 0.0404 | |
|
| 0.0262 | 2.81 | 310 | 0.0394 | |
|
| 0.0239 | 2.9 | 320 | 0.0375 | |
|
| 0.0271 | 2.99 | 330 | 0.0375 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.0 |
|
|
