Downloads · 30 days
0
PHI30512HMAB1H
Litzy619/PHI30512HMAB1H
PHI30512HMAB1H is a machine learning model from Litzy619. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
Access
Public
Updated May 12, 2024
Parameters
3.9B
8.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors7.9 GB · 96%
Parameter types
How the weights are stored.
BF163.8B · 99%
Base models
Model card
From the Hugging Face model README
<!-- 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. -->
PHI30512HMAB1H
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: 0.0701
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 |
|---|---|---|---|
| 5.2344 | 0.09 | 10 | 2.7793 |
| 1.4671 | 0.18 | 20 | 0.5131 |
| 0.3676 | 0.27 | 30 | 2.8781 |
| 0.907 | 0.36 | 40 | 0.2773 |
| 0.2875 | 0.45 | 50 | 0.2421 |
| 0.2486 | 0.54 | 60 | 0.2263 |
| 0.168 | 0.63 | 70 | 0.1595 |
| 0.1505 | 0.73 | 80 | 0.1210 |
| 0.1137 | 0.82 | 90 | 0.1122 |
| 0.1072 | 0.91 | 100 | 0.0915 |
| 0.0906 | 1.0 | 110 | 0.0853 |
| 0.0752 | 1.09 | 120 | 0.0731 |
| 0.0625 | 1.18 | 130 | 0.0723 |
| 0.0649 | 1.27 | 140 | 0.0678 |
| 0.0563 | 1.36 | 150 | 0.0720 |
| 0.0656 | 1.45 | 160 | 0.0662 |
| 0.0638 | 1.54 | 170 | 0.0649 |
| 0.0603 | 1.63 | 180 | 0.0649 |
| 0.0537 | 1.72 | 190 | 0.0626 |
| 0.0638 | 1.81 | 200 | 0.0605 |
| 0.0523 | 1.9 | 210 | 0.0721 |
| 0.0637 | 1.99 | 220 | 0.0634 |
| 0.0384 | 2.08 | 230 | 0.0658 |
| 0.0345 | 2.18 | 240 | 0.0741 |
| 0.0292 | 2.27 | 250 | 0.0753 |
| 0.0323 | 2.36 | 260 | 0.0699 |
| 0.0378 | 2.45 | 270 | 0.0669 |
| 0.0304 | 2.54 | 280 | 0.0712 |
| 0.032 | 2.63 | 290 | 0.0713 |
| 0.0351 | 2.72 | 300 | 0.0706 |
| 0.0388 | 2.81 | 310 | 0.0706 |
| 0.035 | 2.9 | 320 | 0.0697 |
| 0.0318 | 2.99 | 330 | 0.0701 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.0
- License
- mit
- Created
- May 12, 2024
- Updated
- May 12, 2024