|
--- |
|
license: gemma |
|
base_model: google/gemma-2b |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: G0428HMA16 |
|
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. --> |
|
|
|
# G0428HMA16 |
|
|
|
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: 0.1057 |
|
|
|
## 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: 100 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.6372 | 0.09 | 10 | 1.7096 | |
|
| 1.1238 | 0.18 | 20 | 0.4795 | |
|
| 0.2595 | 0.27 | 30 | 0.1668 | |
|
| 0.155 | 0.36 | 40 | 0.1603 | |
|
| 0.1489 | 0.45 | 50 | 0.1483 | |
|
| 0.1474 | 0.54 | 60 | 0.1499 | |
|
| 0.1479 | 0.63 | 70 | 0.1470 | |
|
| 0.1491 | 0.73 | 80 | 0.1479 | |
|
| 0.1413 | 0.82 | 90 | 0.1486 | |
|
| 0.1448 | 0.91 | 100 | 0.1479 | |
|
| 0.1492 | 1.0 | 110 | 0.1488 | |
|
| 0.1429 | 1.09 | 120 | 0.1485 | |
|
| 0.1447 | 1.18 | 130 | 0.1485 | |
|
| 0.146 | 1.27 | 140 | 0.1473 | |
|
| 0.1478 | 1.36 | 150 | 0.1466 | |
|
| 0.1423 | 1.45 | 160 | 0.1507 | |
|
| 0.1434 | 1.54 | 170 | 0.1435 | |
|
| 0.1392 | 1.63 | 180 | 0.1377 | |
|
| 0.1379 | 1.72 | 190 | 0.1359 | |
|
| 0.1285 | 1.81 | 200 | 0.1294 | |
|
| 0.1271 | 1.9 | 210 | 0.1303 | |
|
| 0.1269 | 1.99 | 220 | 0.1228 | |
|
| 0.1118 | 2.08 | 230 | 0.1210 | |
|
| 0.1144 | 2.18 | 240 | 0.1153 | |
|
| 0.1106 | 2.27 | 250 | 0.1123 | |
|
| 0.1116 | 2.36 | 260 | 0.1155 | |
|
| 0.1158 | 2.45 | 270 | 0.1118 | |
|
| 0.1066 | 2.54 | 280 | 0.1109 | |
|
| 0.0991 | 2.63 | 290 | 0.1098 | |
|
| 0.1016 | 2.72 | 300 | 0.1064 | |
|
| 0.1029 | 2.81 | 310 | 0.1058 | |
|
| 0.1052 | 2.9 | 320 | 0.1057 | |
|
| 0.106 | 2.99 | 330 | 0.1057 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
