metadata
license: mit
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: HuggingFaceH4/zephyr-7b-beta
model-index:
- name: zephyr_instruct_generation
results: []
zephyr_instruct_generation
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: 0.7409
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: 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_steps: 0.03
- training_steps: 250
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.9344 | 0.4 | 10 | 2.3554 |
| 1.822 | 0.8 | 20 | 1.3051 |
| 1.0639 | 1.2 | 30 | 0.9542 |
| 0.9127 | 1.6 | 40 | 0.8368 |
| 0.8577 | 2.0 | 50 | 0.7642 |
| 0.7091 | 2.4 | 60 | 0.7175 |
| 0.7112 | 2.8 | 70 | 0.6682 |
| 0.5931 | 3.2 | 80 | 0.6508 |
| 0.5809 | 3.6 | 90 | 0.6226 |
| 0.5901 | 4.0 | 100 | 0.5975 |
| 0.4863 | 4.4 | 110 | 0.6240 |
| 0.5024 | 4.8 | 120 | 0.5896 |
| 0.4932 | 5.2 | 130 | 0.6103 |
| 0.44 | 5.6 | 140 | 0.6158 |
| 0.4737 | 6.0 | 150 | 0.6168 |
| 0.4191 | 6.4 | 160 | 0.6859 |
| 0.4179 | 6.8 | 170 | 0.6237 |
| 0.3921 | 7.2 | 180 | 0.6603 |
| 0.3694 | 7.6 | 190 | 0.6766 |
| 0.3822 | 8.0 | 200 | 0.6559 |
| 0.3495 | 8.4 | 210 | 0.6989 |
| 0.341 | 8.8 | 220 | 0.6622 |
| 0.333 | 9.2 | 230 | 0.7345 |
| 0.3357 | 9.6 | 240 | 0.7215 |
| 0.3284 | 10.0 | 250 | 0.7409 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0