metadata
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-14B-Instruct
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: coupling_model
results: []
coupling_model
This model is a fine-tuned version of Qwen/Qwen2.5-14B-Instruct on the couplingDetection_finetune_train dataset. It achieves the following results on the evaluation set:
- Loss: 0.4664
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.656 | 0.1736 | 100 | 0.5767 |
| 0.5695 | 0.3472 | 200 | 0.5406 |
| 0.5849 | 0.5208 | 300 | 0.4985 |
| 0.527 | 0.6944 | 400 | 0.4857 |
| 0.4782 | 0.8681 | 500 | 0.4704 |
| 0.4357 | 1.0417 | 600 | 0.4712 |
| 0.4033 | 1.2153 | 700 | 0.4654 |
| 0.4438 | 1.3889 | 800 | 0.4594 |
| 0.4297 | 1.5625 | 900 | 0.4532 |
| 0.4012 | 1.7361 | 1000 | 0.4480 |
| 0.2866 | 1.9097 | 1100 | 0.4806 |
| 0.2718 | 2.0833 | 1200 | 0.4796 |
| 0.3169 | 2.2569 | 1300 | 0.4686 |
| 0.276 | 2.4306 | 1400 | 0.4692 |
| 0.2616 | 2.6042 | 1500 | 0.4693 |
| 0.2545 | 2.7778 | 1600 | 0.4671 |
| 0.2673 | 2.9514 | 1700 | 0.4665 |
Framework versions
- PEFT 0.15.2
- Transformers 4.52.4
- Pytorch 2.7.0+cu128
- Datasets 3.6.0
- Tokenizers 0.21.1