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: 1.1686
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: 8e-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_steps: 0.03
- training_steps: 200
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.048 | 0.77 | 10 | 2.4666 |
2.1245 | 1.54 | 20 | 1.6932 |
1.3178 | 2.31 | 30 | 1.1619 |
1.0341 | 3.08 | 40 | 0.9138 |
0.8234 | 3.85 | 50 | 0.8017 |
0.6618 | 4.62 | 60 | 0.7330 |
0.6123 | 5.38 | 70 | 0.7278 |
0.539 | 6.15 | 80 | 0.6836 |
0.4648 | 6.92 | 90 | 0.7171 |
0.4244 | 7.69 | 100 | 0.7712 |
0.3771 | 8.46 | 110 | 0.8586 |
0.351 | 9.23 | 120 | 0.8261 |
0.3233 | 10.0 | 130 | 0.8205 |
0.2946 | 10.77 | 140 | 0.8420 |
0.289 | 11.54 | 150 | 1.0008 |
0.2604 | 12.31 | 160 | 0.9676 |
0.2615 | 13.08 | 170 | 1.0381 |
0.2557 | 13.85 | 180 | 1.0656 |
0.2514 | 14.62 | 190 | 1.0422 |
0.2357 | 15.38 | 200 | 1.1686 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0