metadata
library_name: transformers
license: mit
base_model: 100daggers/reuters-gpt2-text-gen
tags:
- generated_from_trainer
model-index:
- name: reuters-gpt2-text-gen
results: []
reuters-gpt2-text-gen
This model is a fine-tuned version of 100daggers/reuters-gpt2-text-gen on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.9508
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.2626 | 0.9987 | 286 | 4.4727 |
| 3.7809 | 1.9974 | 572 | 4.2128 |
| 3.5797 | 2.9996 | 859 | 4.0473 |
| 3.2899 | 3.9983 | 1145 | 3.9684 |
| 3.2056 | 4.9935 | 1430 | 3.9508 |
Framework versions
- Transformers 4.44.1
- Pytorch 2.0.1
- Datasets 2.21.0
- Tokenizers 0.19.1