File size: 1,853 Bytes
c5ec788 |
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 |
---
base_model: zake7749/gemma-2-2b-it-chinese-kyara-dpo
library_name: peft
license: gemma
tags:
- generated_from_trainer
model-index:
- name: adl-hw3-finetune-gemma-2-chinese-kyara-6
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. -->
# adl-hw3-finetune-gemma-2-chinese-kyara-6
This model is a fine-tuned version of [zake7749/gemma-2-2b-it-chinese-kyara-dpo](https://huggingface.co/zake7749/gemma-2-2b-it-chinese-kyara-dpo) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1672
## 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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 2.3052 | 0.3556 | 200 | 2.1474 |
| 2.1059 | 0.7111 | 400 | 2.0666 |
| 1.9761 | 1.0667 | 600 | 2.0535 |
| 1.7046 | 1.4222 | 800 | 2.0527 |
| 1.7285 | 1.7778 | 1000 | 2.0320 |
| 1.5505 | 2.1333 | 1200 | 2.1532 |
| 1.3239 | 2.4889 | 1400 | 2.1683 |
| 1.3304 | 2.8444 | 1600 | 2.1672 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.45.1
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3 |