lmind_nq_train600_eval300_v1_recite_qa_gpt2-xl

This model is a fine-tuned version of gpt2-xl on the tyzhu/lmind_nq_train600_eval300_v1_recite_qa dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3660
  • Accuracy: 0.8407

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: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • num_epochs: 20.0

Training results

Training Loss Epoch Step Accuracy Validation Loss
2.4392 0.51 47 0.5938 2.1629
2.2015 1.01 94 0.6250 1.8377
1.5193 1.52 141 0.6524 1.8218
1.405 2.02 188 0.6983 1.2418
0.8654 2.53 235 0.7264 1.0606
0.7158 3.03 282 0.7665 0.7902
0.4185 3.54 329 0.7867 0.6745
0.4556 4.04 376 0.8105 0.5201
0.253 4.55 423 0.8198 0.4699
0.1895 5.05 470 0.8301 0.4096
0.1649 5.56 517 0.8323 0.3918
0.1468 6.06 564 0.8367 0.3684
0.1179 6.57 611 0.8377 0.3645
0.1171 7.08 658 0.8394 0.3590
0.0953 7.58 705 0.8402 0.3537
0.0831 8.09 752 0.8400 0.3615
0.0824 8.59 799 0.8408 0.3527
0.0764 9.1 846 0.8410 0.3586
0.0731 9.6 893 0.8408 0.3589
0.0637 10.11 940 0.3618 0.8412
0.069 10.61 987 0.3563 0.8415
0.0636 11.12 1034 0.3650 0.8409
0.0663 11.62 1081 0.3554 0.8418
0.0636 12.13 1128 0.3675 0.8409
0.0632 12.63 1175 0.3590 0.8418
0.0605 13.14 1222 0.3635 0.8413
0.0621 13.65 1269 0.3558 0.8413
0.0579 14.15 1316 0.3682 0.8410
0.0618 14.66 1363 0.3653 0.8405
0.0552 15.16 1410 0.3661 0.8413
0.0619 15.67 1457 0.3596 0.8416
0.0536 16.17 1504 0.3710 0.8414
0.0602 16.68 1551 0.3609 0.8418
0.054 17.18 1598 0.3759 0.8410
0.0635 17.69 1645 0.3597 0.8414
0.0536 18.19 1692 0.3750 0.8410
0.0588 18.7 1739 0.3684 0.8414
0.0713 19.2 1786 0.3691 0.8411
0.0704 19.71 1833 0.3638 0.8413

Framework versions

  • Transformers 4.34.0
  • Pytorch 2.1.0+cu121
  • Datasets 2.14.5
  • Tokenizers 0.14.1
Downloads last month
10
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for tyzhu/lmind_nq_train600_eval300_v1_recite_qa_gpt2-xl

Finetuned
(46)
this model

Dataset used to train tyzhu/lmind_nq_train600_eval300_v1_recite_qa_gpt2-xl

Evaluation results

  • Accuracy on tyzhu/lmind_nq_train600_eval300_v1_recite_qa
    self-reported
    0.841