Configuration Parsing Warning:In adapter_config.json: "peft.task_type" must be a string
rag_deberta_checkpoints
This model is a fine-tuned version of google/bigbird-roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6086
- Map@3: 0.4102
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: 5e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- 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: linear
- lr_scheduler_warmup_steps: 25
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Map@3 |
|---|---|---|---|---|
| 25.9079 | 1.0 | 62 | 1.6103 | 0.3054 |
| 25.7527 | 2.0 | 124 | 1.6093 | 0.3959 |
| 25.6303 | 3.0 | 186 | 1.6084 | 0.3878 |
| 25.8033 | 4.0 | 248 | 1.6082 | 0.4088 |
| 25.6139 | 5.0 | 310 | 1.6086 | 0.4102 |
Framework versions
- PEFT 0.19.1
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2
- Downloads last month
- 5
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for 23f3003974/rag_deberta_checkpoints
Base model
google/bigbird-roberta-large