metadata
library_name: peft
license: mit
base_model: microsoft/phi-1_5
tags:
- generated_from_trainer
model-index:
- name: phi_1.5_mfr_instruct
results: []
phi_1.5_mfr_instruct
This model is a fine-tuned version of microsoft/phi-1_5 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4011
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: 1e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.31 | 50.0 | 250 | 1.2487 |
| 0.5629 | 100.0 | 500 | 0.4522 |
| 0.2176 | 150.0 | 750 | 0.4012 |
| 0.1899 | 200.0 | 1000 | 0.3989 |
| 0.1832 | 250.0 | 1250 | 0.3996 |
| 0.1801 | 300.0 | 1500 | 0.4006 |
| 0.1788 | 350.0 | 1750 | 0.4011 |
| 0.1784 | 400.0 | 2000 | 0.4011 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0