File size: 3,099 Bytes
652fde7 |
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: Phi0503HMA17
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. -->
# Phi0503HMA17
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: 0.0613
## 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: 60
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 3.8235 | 0.09 | 10 | 0.4394 |
| 0.3037 | 0.18 | 20 | 0.2360 |
| 0.2646 | 0.27 | 30 | 0.2336 |
| 0.2183 | 0.36 | 40 | 0.1923 |
| 0.164 | 0.45 | 50 | 0.2113 |
| 0.2323 | 0.54 | 60 | 0.1157 |
| 0.0954 | 0.63 | 70 | 0.0939 |
| 0.0792 | 0.73 | 80 | 0.0938 |
| 0.0914 | 0.82 | 90 | 0.0814 |
| 0.0757 | 0.91 | 100 | 0.0724 |
| 0.0795 | 1.0 | 110 | 0.0717 |
| 0.0546 | 1.09 | 120 | 0.0677 |
| 0.0535 | 1.18 | 130 | 0.0718 |
| 0.0617 | 1.27 | 140 | 0.0718 |
| 0.0561 | 1.36 | 150 | 0.0765 |
| 0.0632 | 1.45 | 160 | 0.0595 |
| 0.0549 | 1.54 | 170 | 0.0612 |
| 0.0404 | 1.63 | 180 | 0.0521 |
| 0.0353 | 1.72 | 190 | 0.0431 |
| 0.0396 | 1.81 | 200 | 0.0489 |
| 0.0272 | 1.9 | 210 | 0.0543 |
| 0.032 | 1.99 | 220 | 0.0489 |
| 0.0132 | 2.08 | 230 | 0.0512 |
| 0.0101 | 2.18 | 240 | 0.0641 |
| 0.0089 | 2.27 | 250 | 0.0688 |
| 0.0095 | 2.36 | 260 | 0.0623 |
| 0.0072 | 2.45 | 270 | 0.0620 |
| 0.0086 | 2.54 | 280 | 0.0628 |
| 0.0073 | 2.63 | 290 | 0.0624 |
| 0.0068 | 2.72 | 300 | 0.0619 |
| 0.0098 | 2.81 | 310 | 0.0621 |
| 0.0088 | 2.9 | 320 | 0.0618 |
| 0.0074 | 2.99 | 330 | 0.0613 |
### Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.0
|