metadata
license: apache-2.0
base_model: distilbert/distilgpt2
tags:
- generated_from_trainer
datasets:
- eli5_category
metrics:
- bleu
model-index:
- name: distilgpt2-finetuned
results:
- task:
name: Causal Language Modeling
type: text-generation
dataset:
name: eli5_category
type: eli5_category
config: default
split: None
args: default
metrics:
- name: Bleu
type: bleu
value: 0.010587533155110318
distilgpt2-finetuned
This model is a fine-tuned version of distilbert/distilgpt2 on the eli5_category dataset. It achieves the following results on the evaluation set:
- Loss: 3.7703
- Bleu: 0.0106
- Bertscore Precision: 0.1609
- Bertscore Recall: 0.1758
- Bertscore F1: 0.1677
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: 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: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Bertscore Precision | Bertscore Recall | Bertscore F1 |
|---|---|---|---|---|---|---|---|
| 3.8816 | 1.0 | 4000 | 3.7775 | 0.0107 | 0.1607 | 0.1756 | 0.1675 |
| 3.7273 | 2.0 | 8000 | 3.7660 | 0.0107 | 0.1608 | 0.1757 | 0.1676 |
| 3.6125 | 3.0 | 12000 | 3.7703 | 0.0106 | 0.1609 | 0.1758 | 0.1677 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1