File size: 3,096 Bytes
411eadf |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 |
---
license: mit
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
- generated_from_trainer
model-index:
- name: Phi0503MA2
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. -->
# Phi0503MA2
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: 0.0832
## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 3.7578 | 0.09 | 10 | 0.9244 |
| 0.4383 | 0.18 | 20 | 0.1617 |
| 0.1537 | 0.27 | 30 | 0.1389 |
| 0.1365 | 0.36 | 40 | 0.1165 |
| 0.1071 | 0.45 | 50 | 0.0962 |
| 0.1021 | 0.54 | 60 | 0.0964 |
| 0.0866 | 0.63 | 70 | 0.0848 |
| 0.0997 | 0.73 | 80 | 0.0891 |
| 0.08 | 0.82 | 90 | 0.0861 |
| 0.0813 | 0.91 | 100 | 0.0706 |
| 0.0675 | 1.0 | 110 | 0.0656 |
| 0.0626 | 1.09 | 120 | 0.0832 |
| 0.0641 | 1.18 | 130 | 0.0733 |
| 0.0693 | 1.27 | 140 | 0.0679 |
| 0.055 | 1.36 | 150 | 0.0745 |
| 0.0572 | 1.45 | 160 | 0.0603 |
| 0.0487 | 1.54 | 170 | 0.0614 |
| 0.0501 | 1.63 | 180 | 0.0610 |
| 0.0456 | 1.72 | 190 | 0.0660 |
| 0.0496 | 1.81 | 200 | 0.0626 |
| 0.0415 | 1.9 | 210 | 0.0626 |
| 0.0463 | 1.99 | 220 | 0.0663 |
| 0.0241 | 2.08 | 230 | 0.0741 |
| 0.0216 | 2.18 | 240 | 0.0932 |
| 0.0186 | 2.27 | 250 | 0.0979 |
| 0.0166 | 2.36 | 260 | 0.0852 |
| 0.0207 | 2.45 | 270 | 0.0819 |
| 0.0148 | 2.54 | 280 | 0.0857 |
| 0.0189 | 2.63 | 290 | 0.0866 |
| 0.0226 | 2.72 | 300 | 0.0844 |
| 0.0229 | 2.81 | 310 | 0.0841 |
| 0.02 | 2.9 | 320 | 0.0833 |
| 0.0187 | 2.99 | 330 | 0.0832 |
### Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1
|