TinyLlama-de-stage3-dr-v1.0
This model is a fine-tuned version of olemeyer/TinyLlama-de-stage3-v0.6 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2203
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: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 32
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.3181 | 0.4 | 50 | 0.2922 |
0.2778 | 0.8 | 100 | 0.2748 |
0.1822 | 1.2 | 150 | 0.2577 |
0.1272 | 1.6 | 200 | 0.2417 |
0.131 | 2.0 | 250 | 0.2156 |
0.0799 | 2.4 | 300 | 0.2168 |
0.0722 | 2.8 | 350 | 0.2203 |
Framework versions
- Transformers 4.34.1
- Pytorch 1.14.0a0+44dac51
- Datasets 2.14.7
- Tokenizers 0.14.1
- Downloads last month
- 5