metadata
license: mit
base_model: HuggingFaceH4/zephyr-7b-beta
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: results
results: []
results
This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-beta on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6394
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: 2.5e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.367 | 0.35 | 100 | 1.3594 |
| 0.774 | 0.69 | 200 | 0.9080 |
| 0.7724 | 1.04 | 300 | 0.7654 |
| 0.6987 | 1.39 | 400 | 0.7205 |
| 0.6482 | 1.74 | 500 | 0.6828 |
| 0.5204 | 2.08 | 600 | 0.6499 |
| 0.5828 | 2.43 | 700 | 0.6353 |
| 0.522 | 2.78 | 800 | 0.6170 |
| 0.3334 | 3.12 | 900 | 0.6522 |
| 0.3643 | 3.47 | 1000 | 0.6149 |
| 0.3551 | 3.82 | 1100 | 0.6138 |
| 0.3144 | 4.17 | 1200 | 0.6430 |
| 0.3194 | 4.51 | 1300 | 0.6468 |
| 0.3341 | 4.86 | 1400 | 0.6287 |
| 0.2982 | 5.21 | 1500 | 0.6403 |
| 0.3056 | 5.56 | 1600 | 0.6394 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1