<!-- 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
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6773
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 |
---|---|---|---|
1.8056 | 0.1 | 1000 | 1.0889 |
1.0277 | 0.19 | 2000 | 0.9550 |
0.9453 | 0.29 | 3000 | 0.9052 |
0.9026 | 0.39 | 4000 | 0.8735 |
0.8751 | 0.48 | 5000 | 0.8492 |
0.8538 | 0.58 | 6000 | 0.8306 |
0.8381 | 0.68 | 7000 | 0.8125 |
0.8225 | 0.77 | 8000 | 0.8030 |
0.8124 | 0.87 | 9000 | 0.7906 |
0.8 | 0.97 | 10000 | 0.7795 |
0.7895 | 1.07 | 11000 | 0.7724 |
0.78 | 1.16 | 12000 | 0.7617 |
0.7729 | 1.26 | 13000 | 0.7570 |
0.766 | 1.36 | 14000 | 0.7477 |
0.7577 | 1.45 | 15000 | 0.7425 |
0.7524 | 1.55 | 16000 | 0.7354 |
0.7454 | 1.65 | 17000 | 0.7286 |
0.739 | 1.74 | 18000 | 0.7220 |
0.7328 | 1.84 | 19000 | 0.7169 |
0.7276 | 1.94 | 20000 | 0.7095 |
0.7184 | 2.03 | 21000 | 0.7040 |
0.71 | 2.13 | 22000 | 0.6994 |
0.7064 | 2.23 | 23000 | 0.6948 |
0.7017 | 2.32 | 24000 | 0.6904 |
0.6982 | 2.42 | 25000 | 0.6866 |
0.6939 | 2.52 | 26000 | 0.6834 |
0.6909 | 2.61 | 27000 | 0.6811 |
0.6896 | 2.71 | 28000 | 0.6790 |
0.6881 | 2.81 | 29000 | 0.6778 |
0.6873 | 2.91 | 30000 | 0.6773 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3