metadata
license: mit
base_model: cuba6112/orion
tags:
- generated_from_trainer
model-index:
- name: orion
results: []
orion
This model is a fine-tuned version of cuba6112/orion on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.5251
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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0871 | 400 | 3.5515 |
| 1.4611 | 0.1743 | 800 | 3.6262 |
| 1.174 | 0.2614 | 1200 | 3.6164 |
| 1.2082 | 0.3486 | 1600 | 3.6236 |
| 1.4301 | 0.4357 | 2000 | 3.5137 |
| 1.4301 | 0.5229 | 2400 | 3.5171 |
| 1.4987 | 0.6100 | 2800 | 3.5004 |
| 1.518 | 0.6972 | 3200 | 3.4667 |
| 1.5859 | 0.7843 | 3600 | 3.4521 |
| 1.6333 | 0.8715 | 4000 | 3.4452 |
| 1.6333 | 0.9586 | 4400 | 3.4300 |
| 1.6698 | 1.0458 | 4800 | 3.5143 |
| 1.4993 | 1.1329 | 5200 | 3.5234 |
| 1.4858 | 1.2200 | 5600 | 3.5240 |
| 1.4804 | 1.3072 | 6000 | 3.4979 |
| 1.4804 | 1.3943 | 6400 | 3.5131 |
| 1.4814 | 1.4815 | 6800 | 3.5177 |
| 1.478 | 1.5686 | 7200 | 3.4989 |
| 1.5073 | 1.6558 | 7600 | 3.5158 |
| 1.4952 | 1.7429 | 8000 | 3.5145 |
| 1.4952 | 1.8301 | 8400 | 3.4975 |
| 1.5367 | 1.9172 | 8800 | 3.5075 |
| 1.5085 | 2.0044 | 9200 | 3.5058 |
| 1.462 | 2.0915 | 9600 | 3.5352 |
| 1.4378 | 2.1786 | 10000 | 3.5335 |
| 1.4378 | 2.2658 | 10400 | 3.5378 |
| 1.4514 | 2.3529 | 10800 | 3.5383 |
| 1.448 | 2.4401 | 11200 | 3.5369 |
| 1.46 | 2.5272 | 11600 | 3.5361 |
| 1.4722 | 2.6144 | 12000 | 3.5337 |
| 1.4722 | 2.7015 | 12400 | 3.5277 |
| 1.4726 | 2.7887 | 12800 | 3.5284 |
| 1.4829 | 2.8758 | 13200 | 3.5257 |
| 1.4963 | 2.9630 | 13600 | 3.5253 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1