metadata
license: mit
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: microsoft/Phi-3-mini-4k-instruct
model-index:
- name: Phi-3-mini-4k-instruct-retrained_network_corpus
results: []
Phi-3-mini-4k-instruct-retrained_network_corpus
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: 4.3960
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: 0.0002
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.8696 | 1.0 | 112 | 2.0452 |
| 1.3684 | 2.0 | 224 | 2.1041 |
| 0.9071 | 3.0 | 336 | 2.3714 |
| 0.5282 | 4.0 | 448 | 2.8351 |
| 0.2728 | 5.0 | 560 | 3.3396 |
| 0.1308 | 6.0 | 672 | 3.7990 |
| 0.0594 | 7.0 | 784 | 3.9724 |
| 0.032 | 8.0 | 896 | 4.0585 |
| 0.0186 | 9.0 | 1008 | 4.3351 |
| 0.0133 | 10.0 | 1120 | 4.3960 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1