metadata
library_name: peft
license: mit
base_model: HuggingFaceH4/zephyr-7b-beta
tags:
- generated_from_trainer
model-index:
- name: zephyr-7b-beta-009-3000
results: []
zephyr-7b-beta-009-3000
This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-beta on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3982
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: 2e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use paged_adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0228 | 0.8658 | 100 | 1.0526 |
| 0.8533 | 1.7273 | 200 | 0.8571 |
| 0.7221 | 2.5887 | 300 | 0.7287 |
| 0.5558 | 3.4502 | 400 | 0.6356 |
| 0.5201 | 4.3117 | 500 | 0.5765 |
| 0.4312 | 5.1732 | 600 | 0.5131 |
| 0.3982 | 6.0346 | 700 | 0.4702 |
| 0.3737 | 6.9004 | 800 | 0.4441 |
| 0.3465 | 7.7619 | 900 | 0.4242 |
| 0.3485 | 8.6234 | 1000 | 0.4079 |
| 0.2942 | 9.4848 | 1100 | 0.3982 |
Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1