Hugging Face's logo

AlekseyKorshuk
/
dalio-all-io-1.3b-3-epoch

Text Generation
Transformers
PyTorch
opt
Generated from Trainer
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
dalio-all-io-1.3b-3-epoch / README.md
AlekseyKorshuk's picture
AlekseyKorshuk
update model card README.md
f9b4b82
preview code
|
raw
history blame
6.78 kB
metadata
license: other
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: dalio-all-io-1.3b-3-epoch
    results: []

dalio-all-io-1.3b-3-epoch

This model is a fine-tuned version of facebook/opt-1.3b on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.3008
  • Accuracy: 0.0584

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: 3.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.6543 0.03 1 2.6113 0.0513
2.6077 0.07 2 2.6113 0.0513
2.5964 0.1 3 2.5605 0.0519
2.7302 0.14 4 2.5234 0.0526
2.7004 0.17 5 2.5078 0.0529
2.5681 0.21 6 2.4941 0.0532
2.6404 0.24 7 2.4883 0.0534
2.5325 0.28 8 2.4805 0.0536
2.7205 0.31 9 2.4746 0.0536
2.5149 0.34 10 2.4648 0.0533
2.5017 0.38 11 2.4512 0.0535
2.7026 0.41 12 2.4395 0.0539
2.5259 0.45 13 2.4316 0.0543
2.563 0.48 14 2.4219 0.0546
2.5679 0.52 15 2.4141 0.0550
2.3701 0.55 16 2.4082 0.0551
2.4739 0.59 17 2.4082 0.0551
2.481 0.62 18 2.4023 0.0548
2.5795 0.66 19 2.3945 0.0549
2.4902 0.69 20 2.3867 0.0549
2.4509 0.72 21 2.3809 0.0551
2.6052 0.76 22 2.3730 0.0553
2.3323 0.79 23 2.3633 0.0555
2.5994 0.83 24 2.3555 0.0556
2.3347 0.86 25 2.3477 0.0556
2.421 0.9 26 2.3398 0.0559
2.5337 0.93 27 2.3359 0.0560
2.4102 0.97 28 2.3320 0.0563
2.4309 1.0 29 2.3262 0.0564
1.9305 1.03 30 2.3223 0.0564
1.8601 1.07 31 2.3203 0.0567
1.8682 1.1 32 2.3281 0.0564
1.8657 1.14 33 2.3535 0.0564
2.063 1.17 34 2.3398 0.0567
1.6443 1.21 35 2.3242 0.0568
1.7592 1.24 36 2.3164 0.0569
1.8981 1.28 37 2.3105 0.0569
1.9379 1.31 38 2.3047 0.0573
1.6008 1.34 39 2.3027 0.0574
1.595 1.38 40 2.3027 0.0575
1.7096 1.41 41 2.3027 0.0575
1.7245 1.45 42 2.3027 0.0576
1.795 1.48 43 2.3008 0.0577
1.7241 1.52 44 2.3008 0.0576
1.6356 1.55 45 2.2988 0.0576
1.77 1.59 46 2.2969 0.0576
1.6675 1.62 47 2.2930 0.0577
1.6929 1.66 48 2.2910 0.0577
1.6635 1.69 49 2.2910 0.0576
1.6093 1.72 50 2.2910 0.0578
1.7362 1.76 51 2.2891 0.0580
1.7015 1.79 52 2.2852 0.0581
1.9515 1.83 53 2.2812 0.0582
1.6494 1.86 54 2.2773 0.0580
1.7522 1.9 55 2.2734 0.0580
1.7369 1.93 56 2.2676 0.0581
1.6528 1.97 57 2.2637 0.0581
1.51 2.0 58 2.2617 0.0583
1.4579 2.03 59 2.2637 0.0585
1.2645 2.07 60 2.2695 0.0585
1.2424 2.1 61 2.2773 0.0584
1.2117 2.14 62 2.2891 0.0584
1.4059 2.17 63 2.3008 0.0580
1.328 2.21 64 2.3145 0.0581
1.3436 2.24 65 2.3281 0.0580
1.389 2.28 66 2.3379 0.0580
1.2127 2.31 67 2.3398 0.0580
1.3645 2.34 68 2.3418 0.0581
1.3389 2.38 69 2.3379 0.0581
1.2549 2.41 70 2.3320 0.0581
1.2193 2.45 71 2.3281 0.0582
1.3617 2.48 72 2.3223 0.0583
1.2336 2.52 73 2.3184 0.0583
1.179 2.55 74 2.3145 0.0583
1.2468 2.59 75 2.3125 0.0583
1.3325 2.62 76 2.3086 0.0583
1.1471 2.66 77 2.3066 0.0583
1.3123 2.69 78 2.3066 0.0583
1.3285 2.72 79 2.3047 0.0585
1.3232 2.76 80 2.3027 0.0584
1.1228 2.79 81 2.3027 0.0584
1.3524 2.83 82 2.3027 0.0584
1.2042 2.86 83 2.3027 0.0583
1.3588 2.9 84 2.3008 0.0583
1.2982 2.93 85 2.3008 0.0584
1.4373 2.97 86 2.3008 0.0585
1.3562 3.0 87 2.3008 0.0584

Framework versions

  • Transformers 4.25.0.dev0
  • Pytorch 1.12.1+cu113
  • Datasets 2.3.2
  • Tokenizers 0.12.1