metadata
license: mit
base_model: ai-forever/ruElectra-small
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: electra128
results: []
electra128
This model is a fine-tuned version of ai-forever/ruElectra-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2154
- Accuracy: 0.5134
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: 1e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4844 | 1.0 | 1407 | 1.4000 | 0.4018 |
| 1.3624 | 2.0 | 2814 | 1.3090 | 0.4629 |
| 1.2965 | 3.0 | 4221 | 1.2579 | 0.4897 |
| 1.2577 | 4.0 | 5628 | 1.2271 | 0.5062 |
| 1.2402 | 5.0 | 7035 | 1.2154 | 0.5134 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1