metadata
base_model: microsoft/Phi-3.5-mini-instruct
library_name: peft
license: mit
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: guru1984
results: []
guru1984
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: 1.9331
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.0802 | 0.0786 | 50 | 3.3923 |
| 2.6833 | 0.1572 | 100 | 2.3272 |
| 2.1534 | 0.2358 | 150 | 2.1625 |
| 2.0476 | 0.3145 | 200 | 2.1005 |
| 1.9929 | 0.3931 | 250 | 2.0717 |
| 1.974 | 0.4717 | 300 | 2.0509 |
| 1.9854 | 0.5503 | 350 | 2.0435 |
| 2.0109 | 0.6289 | 400 | 2.0358 |
| 1.9939 | 0.7075 | 450 | 2.0285 |
| 1.9672 | 0.7862 | 500 | 2.0150 |
| 1.9565 | 0.8648 | 550 | 2.0115 |
| 1.9706 | 0.9434 | 600 | 2.0062 |
| 1.9352 | 1.0220 | 650 | 1.9949 |
| 1.8647 | 1.1006 | 700 | 1.9930 |
| 1.9109 | 1.1792 | 750 | 1.9840 |
| 1.885 | 1.2579 | 800 | 1.9806 |
| 1.8864 | 1.3365 | 850 | 1.9878 |
| 1.8931 | 1.4151 | 900 | 1.9825 |
| 1.8599 | 1.4937 | 950 | 1.9755 |
| 1.9019 | 1.5723 | 1000 | 1.9695 |
| 1.9081 | 1.6509 | 1050 | 1.9599 |
| 1.8736 | 1.7296 | 1100 | 1.9583 |
| 1.8939 | 1.8082 | 1150 | 1.9567 |
| 1.8867 | 1.8868 | 1200 | 1.9534 |
| 1.8875 | 1.9654 | 1250 | 1.9479 |
| 1.8677 | 2.0440 | 1300 | 1.9498 |
| 1.8152 | 2.1226 | 1350 | 1.9527 |
| 1.8573 | 2.2013 | 1400 | 1.9523 |
| 1.8433 | 2.2799 | 1450 | 1.9441 |
| 1.828 | 2.3585 | 1500 | 1.9455 |
| 1.8298 | 2.4371 | 1550 | 1.9423 |
| 1.8258 | 2.5157 | 1600 | 1.9376 |
| 1.8314 | 2.5943 | 1650 | 1.9373 |
| 1.8436 | 2.6730 | 1700 | 1.9394 |
| 1.8253 | 2.7516 | 1750 | 1.9373 |
| 1.8055 | 2.8302 | 1800 | 1.9365 |
| 1.8207 | 2.9088 | 1850 | 1.9342 |
| 1.8514 | 2.9874 | 1900 | 1.9331 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1