metadata
license: other
base_model: facebook/opt-350m
tags:
- generated_from_trainer
model-index:
- name: sw-opt-350m
results: []
sw-opt-350m
This model is a fine-tuned version of facebook/opt-350m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.6650
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: 0.0004
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 1024
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 200
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.2094 | 0.1147 | 200 | 5.0391 |
| 4.4151 | 0.2294 | 400 | 4.3766 |
| 3.9865 | 0.3441 | 600 | 4.1046 |
| 3.7852 | 0.4588 | 800 | 3.9390 |
| 3.6542 | 0.5735 | 1000 | 3.8254 |
| 3.5559 | 0.6882 | 1200 | 3.7421 |
| 3.4816 | 0.8029 | 1400 | 3.6885 |
| 3.447 | 0.9176 | 1600 | 3.6650 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.2.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1