File size: 1,887 Bytes
ed19e81 |
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: other
library_name: peft
tags:
- generated_from_trainer
base_model: google/gemma-2b-it
model-index:
- name: outputs
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. -->
# outputs
This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0962
## 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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 10
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 3.5696 | 0.93 | 7 | 2.3790 |
| 1.7335 | 2.0 | 15 | 1.6733 |
| 1.5583 | 2.93 | 22 | 1.5446 |
| 1.1347 | 4.0 | 30 | 1.5097 |
| 1.0502 | 4.93 | 37 | 1.5663 |
| 0.7217 | 6.0 | 45 | 1.6769 |
| 0.6408 | 6.93 | 52 | 1.8386 |
| 0.43 | 8.0 | 60 | 1.9674 |
| 0.391 | 8.93 | 67 | 2.0831 |
| 0.2893 | 9.33 | 70 | 2.0962 |
### Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2 |