<!-- 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. -->
tiny-mlm-glue-sst2-from-scratch-custom-tokenizer-target-glue-mnli
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-sst2-from-scratch-custom-tokenizer on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0169
- Accuracy: 0.4786
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0989 | 0.04 | 500 | 1.0972 | 0.3728 |
| 1.0902 | 0.08 | 1000 | 1.0817 | 0.3793 |
| 1.0758 | 0.12 | 1500 | 1.0706 | 0.4141 |
| 1.0575 | 0.16 | 2000 | 1.0509 | 0.4302 |
| 1.047 | 0.2 | 2500 | 1.0374 | 0.4488 |
| 1.0395 | 0.24 | 3000 | 1.0324 | 0.4532 |
| 1.0332 | 0.29 | 3500 | 1.0350 | 0.4408 |
| 1.0226 | 0.33 | 4000 | 1.0232 | 0.4656 |
| 1.0154 | 0.37 | 4500 | 1.0191 | 0.4740 |
| 1.013 | 0.41 | 5000 | 1.0169 | 0.4786 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2
