metadata
base_model: microsoft/Phi-3.5-mini-instruct
library_name: peft
license: mit
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: Phi-3.5-MultiCap-tool-lora
results: []
Phi-3.5-MultiCap-tool-lora
This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4902
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7255 | 0.2256 | 50 | 0.7138 |
| 0.4783 | 0.4512 | 100 | 0.4788 |
| 0.4616 | 0.6768 | 150 | 0.4543 |
| 0.4794 | 0.9024 | 200 | 0.4437 |
| 0.4174 | 1.1280 | 250 | 0.4357 |
| 0.4097 | 1.3536 | 300 | 0.4310 |
| 0.3829 | 1.5792 | 350 | 0.4280 |
| 0.4358 | 1.8049 | 400 | 0.4264 |
| 0.4013 | 2.0305 | 450 | 0.4261 |
| 0.3685 | 2.2561 | 500 | 0.4268 |
| 0.3823 | 2.4817 | 550 | 0.4276 |
| 0.401 | 2.7073 | 600 | 0.4294 |
| 0.3975 | 2.9329 | 650 | 0.4310 |
| 0.4012 | 3.1585 | 700 | 0.4373 |
| 0.3497 | 3.3841 | 750 | 0.4401 |
| 0.3613 | 3.6097 | 800 | 0.4456 |
| 0.3649 | 3.8353 | 850 | 0.4522 |
| 0.3384 | 4.0609 | 900 | 0.4575 |
| 0.3241 | 4.2865 | 950 | 0.4628 |
| 0.322 | 4.5121 | 1000 | 0.4662 |
| 0.3397 | 4.7377 | 1050 | 0.4720 |
| 0.3228 | 4.9633 | 1100 | 0.4788 |
| 0.3391 | 5.1889 | 1150 | 0.4820 |
| 0.3369 | 5.4146 | 1200 | 0.4861 |
| 0.3424 | 5.6402 | 1250 | 0.4873 |
| 0.3302 | 5.8658 | 1300 | 0.4902 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.4.1+cu124
- Datasets 3.0.0
- Tokenizers 0.19.1