metadata
license: mit
base_model: HuggingFaceH4/zephyr-7b-beta
tags:
- generated_from_trainer
model-index:
- name: non-qa-sft-zephyr-7b-beta-v1
results: []
non-qa-sft-zephyr-7b-beta-v1
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: 0.5768
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3178 | 0.03 | 50 | 1.0767 |
| 0.7765 | 0.07 | 100 | 0.7130 |
| 0.6491 | 0.1 | 150 | 0.6840 |
| 0.6441 | 0.14 | 200 | 0.6829 |
| 0.701 | 0.17 | 250 | 0.6642 |
| 0.6936 | 0.21 | 300 | 0.6427 |
| 0.6538 | 0.24 | 350 | 0.6175 |
| 0.5927 | 0.27 | 400 | 0.6139 |
| 0.6709 | 0.31 | 450 | 0.6129 |
| 0.5961 | 0.34 | 500 | 0.6078 |
| 0.6161 | 0.38 | 550 | 0.5956 |
| 0.5999 | 0.41 | 600 | 0.5938 |
| 0.6248 | 0.44 | 650 | 0.5824 |
| 0.6494 | 0.48 | 700 | 0.5806 |
| 0.6259 | 0.51 | 750 | 0.5767 |
| 0.557 | 0.55 | 800 | 0.5762 |
| 0.6215 | 0.58 | 850 | 0.5777 |
| 0.5986 | 0.62 | 900 | 0.5770 |
| 0.6224 | 0.65 | 950 | 0.5767 |
| 0.6058 | 0.68 | 1000 | 0.5768 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0
- Datasets 2.15.0
- Tokenizers 0.15.0