|
--- |
|
license: gemma |
|
base_model: google/gemma-2b |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: G0521HMA26H2 |
|
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. --> |
|
|
|
# G0521HMA26H2 |
|
|
|
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1233 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 16 |
|
- total_train_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine_with_restarts |
|
- lr_scheduler_warmup_steps: 80 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.7533 | 0.09 | 10 | 1.4230 | |
|
| 1.0512 | 0.18 | 20 | 0.5986 | |
|
| 0.3803 | 0.27 | 30 | 0.1989 | |
|
| 0.1564 | 0.36 | 40 | 0.1744 | |
|
| 0.147 | 0.45 | 50 | 0.2292 | |
|
| 0.1479 | 0.54 | 60 | 0.2788 | |
|
| 0.1448 | 0.63 | 70 | 0.1496 | |
|
| 0.1444 | 0.73 | 80 | 0.1560 | |
|
| 0.1358 | 0.82 | 90 | 0.1688 | |
|
| 0.1287 | 0.91 | 100 | 0.1792 | |
|
| 0.1314 | 1.0 | 110 | 0.1689 | |
|
| 0.1221 | 1.09 | 120 | 0.1936 | |
|
| 0.111 | 1.18 | 130 | 0.1777 | |
|
| 0.1187 | 1.27 | 140 | 0.1412 | |
|
| 0.1216 | 1.36 | 150 | 0.2297 | |
|
| 0.1161 | 1.45 | 160 | 0.1417 | |
|
| 0.1134 | 1.54 | 170 | 0.1235 | |
|
| 0.1096 | 1.63 | 180 | 0.1318 | |
|
| 0.1106 | 1.72 | 190 | 0.1812 | |
|
| 0.1113 | 1.81 | 200 | 0.1179 | |
|
| 0.1089 | 1.9 | 210 | 0.1297 | |
|
| 0.109 | 1.99 | 220 | 0.1219 | |
|
| 0.0944 | 2.08 | 230 | 0.1428 | |
|
| 0.0972 | 2.18 | 240 | 0.1483 | |
|
| 0.089 | 2.27 | 250 | 0.1814 | |
|
| 0.0906 | 2.36 | 260 | 0.2539 | |
|
| 0.0936 | 2.45 | 270 | 0.2099 | |
|
| 0.0864 | 2.54 | 280 | 0.2426 | |
|
| 0.0835 | 2.63 | 290 | 0.1394 | |
|
| 0.0865 | 2.72 | 300 | 0.1288 | |
|
| 0.0923 | 2.81 | 310 | 0.1242 | |
|
| 0.0898 | 2.9 | 320 | 0.1233 | |
|
| 0.0923 | 2.99 | 330 | 0.1233 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.0 |
|
|
