metadata
license: mit
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: microsoft/Phi-3.5-mini-instruct
model-index:
- name: kangoroo-no-subgraph-phi-3-mini-LoRA
results: []
kangoroo-no-subgraph-phi-3-mini-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: 0.2338
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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 |
|---|---|---|---|
| 1.4504 | 0.2892 | 100 | 1.0894 |
| 0.8508 | 0.5785 | 200 | 0.6406 |
| 0.5328 | 0.8677 | 300 | 0.4493 |
| 0.4011 | 1.1569 | 400 | 0.3575 |
| 0.3328 | 1.4461 | 500 | 0.3085 |
| 0.291 | 1.7354 | 600 | 0.2769 |
| 0.2711 | 2.0246 | 700 | 0.2585 |
| 0.2494 | 2.3138 | 800 | 0.2453 |
| 0.2361 | 2.6030 | 900 | 0.2378 |
| 0.2348 | 2.8923 | 1000 | 0.2338 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1