metadata
license: other
tags:
- generated_from_trainer
datasets:
- AlekseyKorshuk/dalio-handwritten-io
metrics:
- accuracy
model-index:
- name: test-clm
results:
- task:
name: Causal Language Modeling
type: text-generation
dataset:
name: AlekseyKorshuk/dalio-handwritten-io
type: AlekseyKorshuk/dalio-handwritten-io
metrics:
- name: Accuracy
type: accuracy
value: 0.06414321574844761
test-clm
This model is a fine-tuned version of facebook/opt-1.3b on the AlekseyKorshuk/dalio-handwritten-io dataset. It achieves the following results on the evaluation set:
- Loss: 2.5547
- Accuracy: 0.0641
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: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 5.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
2.9219 | 0.1 | 1 | 2.6484 | 0.0529 |
2.6938 | 0.2 | 2 | 2.6484 | 0.0529 |
2.6365 | 0.3 | 3 | 2.5508 | 0.0560 |
2.5088 | 0.4 | 4 | 2.5332 | 0.0563 |
2.7297 | 0.5 | 5 | 2.5176 | 0.0567 |
2.9702 | 0.6 | 6 | 2.4941 | 0.0572 |
2.729 | 0.7 | 7 | 2.4883 | 0.0568 |
2.6172 | 0.8 | 8 | 2.4785 | 0.0578 |
2.6428 | 0.9 | 9 | 2.4590 | 0.0581 |
2.5681 | 1.0 | 10 | 2.4355 | 0.0590 |
2.1885 | 1.1 | 11 | 2.4238 | 0.0587 |
1.981 | 1.2 | 12 | 2.4219 | 0.0587 |
1.8673 | 1.3 | 13 | 2.4180 | 0.0591 |
1.7321 | 1.4 | 14 | 2.4180 | 0.0596 |
1.6355 | 1.5 | 15 | 2.4180 | 0.0601 |
1.7758 | 1.6 | 16 | 2.4199 | 0.0602 |
2.0162 | 1.7 | 17 | 2.4082 | 0.0605 |
1.8037 | 1.8 | 18 | 2.3965 | 0.0605 |
1.7204 | 1.9 | 19 | 2.375 | 0.0608 |
1.7831 | 2.0 | 20 | 2.3574 | 0.0609 |
1.299 | 2.1 | 21 | 2.3496 | 0.0616 |
1.4463 | 2.2 | 22 | 2.3496 | 0.0620 |
1.1733 | 2.3 | 23 | 2.3652 | 0.0617 |
1.1142 | 2.4 | 24 | 2.3887 | 0.0626 |
1.3107 | 2.5 | 25 | 2.4219 | 0.0627 |
1.011 | 2.6 | 26 | 2.4551 | 0.0622 |
1.3403 | 2.7 | 27 | 2.4766 | 0.0616 |
1.3108 | 2.8 | 28 | 2.4766 | 0.0616 |
1.0076 | 2.9 | 29 | 2.4609 | 0.0619 |
0.8656 | 3.0 | 30 | 2.4512 | 0.0624 |
0.6635 | 3.1 | 31 | 2.4512 | 0.0628 |
0.9996 | 3.2 | 32 | 2.4434 | 0.0635 |
0.9029 | 3.3 | 33 | 2.4473 | 0.0637 |
0.8329 | 3.4 | 34 | 2.4551 | 0.0637 |
0.8012 | 3.5 | 35 | 2.4648 | 0.0639 |
0.5814 | 3.6 | 36 | 2.4902 | 0.0640 |
1.0688 | 3.7 | 37 | 2.5098 | 0.0638 |
0.8688 | 3.8 | 38 | 2.5176 | 0.0635 |
0.7341 | 3.9 | 39 | 2.5195 | 0.0638 |
0.7102 | 4.0 | 40 | 2.5195 | 0.0640 |
0.7079 | 4.1 | 41 | 2.5195 | 0.0641 |
0.7656 | 4.2 | 42 | 2.5195 | 0.0643 |
0.6377 | 4.3 | 43 | 2.5273 | 0.0645 |
0.5898 | 4.4 | 44 | 2.5352 | 0.0641 |
0.5958 | 4.5 | 45 | 2.5430 | 0.0641 |
0.7048 | 4.6 | 46 | 2.5488 | 0.0640 |
0.5435 | 4.7 | 47 | 2.5527 | 0.0641 |
0.4769 | 4.8 | 48 | 2.5527 | 0.0640 |
0.6583 | 4.9 | 49 | 2.5547 | 0.0642 |
0.7168 | 5.0 | 50 | 2.5547 | 0.0641 |
Framework versions
- Transformers 4.25.0.dev0
- Pytorch 1.12.1+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1