debiased
This model is a fine-tuned version of samzirbo/gendered on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2305
- Bleu: 43.086
- Meteor: 0.6836
- Chrf++: 61.9553
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- training_steps: 10000
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Meteor | Chrf++ |
|---|---|---|---|---|---|---|
| 1.3358 | 3.2 | 2000 | 1.1980 | 43.4198 | 0.6874 | 62.3558 |
| 1.2174 | 6.4 | 4000 | 1.2181 | 43.3333 | 0.6844 | 62.1229 |
| 1.1529 | 9.6 | 6000 | 1.2246 | 43.222 | 0.6846 | 62.054 |
| 1.1161 | 12.8 | 8000 | 1.2291 | 43.1386 | 0.6836 | 61.9676 |
| 1.1026 | 16.0 | 10000 | 1.2305 | 43.086 | 0.6836 | 61.9553 |
Framework versions
- Transformers 4.38.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.15.2
- Downloads last month
- 116
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.