metadata
license: mit
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: microsoft/Phi-3-mini-4k-instruct
datasets:
- generator
model-index:
- name: cls_headline_phi3_v1
results: []
cls_headline_phi3_v1
This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.2738
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3433 | 0.2395 | 20 | 0.3262 |
| 0.3028 | 0.4790 | 40 | 0.3021 |
| 0.2909 | 0.7186 | 60 | 0.2903 |
| 0.2877 | 0.9581 | 80 | 0.2854 |
| 0.2566 | 1.1976 | 100 | 0.2814 |
| 0.2565 | 1.4371 | 120 | 0.2780 |
| 0.2524 | 1.6766 | 140 | 0.2774 |
| 0.2617 | 1.9162 | 160 | 0.2738 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1