<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert_add_GLUE_Experiment_logit_kd_pretrain_sst2
This model is a fine-tuned version of gokuls/distilbert_add_pre-training-complete on the GLUE SST2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.7266
- Accuracy: 0.8085
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: 256
- eval_batch_size: 256
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.9277 | 1.0 | 264 | 0.7266 | 0.8085 |
0.4581 | 2.0 | 528 | 0.9527 | 0.7844 |
0.3282 | 3.0 | 792 | 0.8676 | 0.8142 |
0.2532 | 4.0 | 1056 | 0.7918 | 0.8039 |
0.1926 | 5.0 | 1320 | 0.8852 | 0.7982 |
0.1573 | 6.0 | 1584 | 1.0020 | 0.7947 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2