gec-flan-t5-xxl
This model is a fine-tuned version of google/flan-t5-xxl on 512duncanl/c4_200m_cleaned_365k, a cleaned subset of Google's C4 200M. It achieves the following results on the evaluation set:
- Loss: 0.1853
- F0.5: 0.4088
- Precision: 0.4237
- Recall: 0.3584
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training was conducted on 4x H100 SXM 80GB
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.000008
- train_batch_size: 18
- eval_batch_size: 18
- seed: 42
- distributed_type: multi-GPU
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant_with_warmup
- num_epochs: 2
- warmup_ratio: 0.1
- weight_decay: 0.001
Training results
| Training Loss | Epoch | Step | Validation Loss | F0.5 | Precision | Recall |
|---|---|---|---|---|---|---|
| 0.2111 | 1.0 | 5000 | 0.1909 | 0.3982 | 0.4145 | 0.3441 |
| 0.1983 | 2.0 | 10000 | 0.1853 | 0.4088 | 0.4237 | 0.3584 |
Framework versions
- Transformers 4.53.3
- Pytorch 2.7.1+cu128
- Datasets 4.0.0
- Tokenizers 0.21.2
- Downloads last month
- 45
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support