File size: 1,787 Bytes
77ed0bc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
license: mit
library_name: peft
tags:
- trl
- kto
- generated_from_trainer
base_model: HuggingFaceH4/zephyr-7b-beta
model-index:
- name: WeniGPT-Agents-Zephyr-1.0.17-KTO
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# WeniGPT-Agents-Zephyr-1.0.17-KTO
This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3619
- Eval/rewards/chosen: 1.7854
- Eval/logps/chosen: -265.8084
- Eval/rewards/rejected: -3.3369
- Eval/logps/rejected: -297.7391
- Eval/rewards/margins: 5.1223
- Eval/kl: 3.9467
## 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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 145
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.4039 | 0.34 | 50 | 0.3767 | 0.7345 |
| 0.302 | 0.68 | 100 | 0.3619 | 3.9467 |
### Framework versions
- PEFT 0.10.0
- Transformers 4.39.1
- Pytorch 2.1.0+cu118
- Datasets 2.18.0
- Tokenizers 0.15.2 |