metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: EleutherAI/pythia-70m
model-index:
- name: polish_wikipedia_model
results: []
polish_wikipedia_model
This model is a fine-tuned version of EleutherAI/pythia-70m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0404
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 133 | 0.4825 |
| No log | 2.0 | 266 | 0.4049 |
| No log | 3.0 | 399 | 0.4124 |
| 0.4774 | 4.0 | 532 | 0.2849 |
| 0.4774 | 5.0 | 665 | 0.2543 |
| 0.4774 | 6.0 | 798 | 0.1860 |
| 0.4774 | 7.0 | 931 | 0.1209 |
| 0.2561 | 8.0 | 1064 | 0.0836 |
| 0.2561 | 9.0 | 1197 | 0.0530 |
| 0.2561 | 10.0 | 1330 | 0.0404 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1