<!-- 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. -->
mini_molformer_gsf_3epochs_512width
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6349
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: 0.0006
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.4417 | 0.1 | 1000 | 1.0328 |
| 0.9742 | 0.19 | 2000 | 0.9153 |
| 0.9039 | 0.29 | 3000 | 0.8725 |
| 0.8669 | 0.39 | 4000 | 0.8445 |
| 0.8399 | 0.48 | 5000 | 0.8172 |
| 0.8205 | 0.58 | 6000 | 0.8014 |
| 0.8022 | 0.68 | 7000 | 0.7837 |
| 0.7887 | 0.77 | 8000 | 0.7754 |
| 0.776 | 0.87 | 9000 | 0.7617 |
| 0.7648 | 0.97 | 10000 | 0.7493 |
| 0.751 | 1.07 | 11000 | 0.7413 |
| 0.7412 | 1.16 | 12000 | 0.7340 |
| 0.7332 | 1.26 | 13000 | 0.7229 |
| 0.725 | 1.36 | 14000 | 0.7170 |
| 0.7176 | 1.45 | 15000 | 0.7073 |
| 0.7097 | 1.55 | 16000 | 0.7007 |
| 0.7029 | 1.65 | 17000 | 0.6921 |
| 0.6953 | 1.74 | 18000 | 0.6850 |
| 0.6891 | 1.84 | 19000 | 0.6786 |
| 0.6817 | 1.94 | 20000 | 0.6733 |
| 0.6709 | 2.03 | 21000 | 0.6648 |
| 0.6582 | 2.13 | 22000 | 0.6594 |
| 0.6539 | 2.23 | 23000 | 0.6541 |
| 0.6486 | 2.32 | 24000 | 0.6496 |
| 0.6437 | 2.42 | 25000 | 0.6452 |
| 0.6411 | 2.52 | 26000 | 0.6417 |
| 0.6379 | 2.61 | 27000 | 0.6386 |
| 0.6347 | 2.71 | 28000 | 0.6367 |
| 0.6338 | 2.81 | 29000 | 0.6355 |
| 0.6326 | 2.91 | 30000 | 0.6349 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
