Hugging Face's logo

Jellywibble
/
CHAI_alignment_reward_model

Text Classification
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
CHAI_alignment_reward_model / README.md
Jellywibble's picture
Jellywibble
Jellywibble/CHAI_alignment_reward_model
5aef637 verified
preview code
|
raw
history blame
8.37 kB
metadata
license: mit
base_model: gpt2
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: output
    results: []

output

This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5590
  • Accuracy: 0.7005

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: 1e-05
  • train_batch_size: 16
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6976 0.0268 250 0.6614 0.6728
0.6155 0.0537 500 0.5858 0.6811
0.5869 0.0805 750 0.5820 0.6856
0.5911 0.1073 1000 0.5843 0.6811
0.5788 0.1341 1250 0.5750 0.6790
0.5913 0.1610 1500 0.5810 0.6864
0.5712 0.1878 1750 0.5731 0.6892
0.5793 0.2146 2000 0.5717 0.6882
0.5788 0.2415 2250 0.5868 0.6838
0.5802 0.2683 2500 0.5653 0.6942
0.583 0.2951 2750 0.5631 0.6984
0.5762 0.3220 3000 0.5654 0.6916
0.5678 0.3488 3250 0.5635 0.6906
0.5679 0.3756 3500 0.5706 0.6838
0.56 0.4024 3750 0.5661 0.6932
0.562 0.4293 4000 0.5994 0.6885
0.5861 0.4561 4250 0.5659 0.6979
0.5845 0.4829 4500 0.5631 0.6992
0.5665 0.5098 4750 0.5621 0.6987
0.5795 0.5366 5000 0.5698 0.6934
0.5722 0.5634 5250 0.5615 0.6895
0.5765 0.5903 5500 0.5610 0.7010
0.5627 0.6171 5750 0.5594 0.6932
0.5761 0.6439 6000 0.5581 0.6997
0.5682 0.6707 6250 0.5693 0.6856
0.566 0.6976 6500 0.5634 0.6895
0.5628 0.7244 6750 0.5594 0.7026
0.5739 0.7512 7000 0.5634 0.6926
0.5762 0.7781 7250 0.5593 0.7015
0.572 0.8049 7500 0.5612 0.6853
0.5657 0.8317 7750 0.5593 0.6974
0.5665 0.8586 8000 0.5614 0.6916
0.578 0.8854 8250 0.5600 0.6995
0.571 0.9122 8500 0.5635 0.6934
0.5703 0.9390 8750 0.5628 0.7052
0.5801 0.9659 9000 0.5582 0.7010
0.5691 0.9927 9250 0.5673 0.6958
0.551 1.0195 9500 0.5631 0.6913
0.5625 1.0464 9750 0.5583 0.6987
0.5679 1.0732 10000 0.5633 0.7015
0.5693 1.1000 10250 0.5590 0.6934
0.5649 1.1269 10500 0.5580 0.6966
0.5558 1.1537 10750 0.5661 0.6879
0.5674 1.1805 11000 0.5595 0.7026
0.5507 1.2073 11250 0.5594 0.7015
0.5656 1.2342 11500 0.5592 0.6976
0.5696 1.2610 11750 0.5604 0.6926
0.5605 1.2878 12000 0.5618 0.7026
0.5572 1.3147 12250 0.5649 0.7000
0.5553 1.3415 12500 0.5621 0.6984
0.546 1.3683 12750 0.5630 0.6966
0.5614 1.3951 13000 0.5605 0.6955
0.5635 1.4220 13250 0.5587 0.6971
0.5561 1.4488 13500 0.5647 0.6947
0.5634 1.4756 13750 0.5607 0.6995
0.5585 1.5025 14000 0.5577 0.7023
0.5599 1.5293 14250 0.5740 0.6788
0.5697 1.5561 14500 0.5570 0.7023
0.5453 1.5830 14750 0.5624 0.6921
0.5642 1.6098 15000 0.5687 0.6864
0.5692 1.6366 15250 0.5643 0.6924
0.558 1.6634 15500 0.5625 0.6961
0.5465 1.6903 15750 0.5627 0.6997
0.5744 1.7171 16000 0.5594 0.6992
0.5683 1.7439 16250 0.5577 0.6961
0.5638 1.7708 16500 0.5579 0.6961
0.5512 1.7976 16750 0.5613 0.6945
0.5652 1.8244 17000 0.5596 0.6987
0.5771 1.8513 17250 0.5575 0.6997
0.5624 1.8781 17500 0.5628 0.6971
0.5719 1.9049 17750 0.5575 0.6937
0.5577 1.9317 18000 0.5686 0.6895
0.5599 1.9586 18250 0.5632 0.6981
0.5622 1.9854 18500 0.5574 0.7008
0.56 2.0122 18750 0.5577 0.7008
0.5447 2.0391 19000 0.5590 0.7036
0.5599 2.0659 19250 0.5604 0.7005
0.5512 2.0927 19500 0.5584 0.7000
0.549 2.1196 19750 0.5593 0.6987
0.5485 2.1464 20000 0.5680 0.6947
0.5528 2.1732 20250 0.5619 0.6955
0.5549 2.2000 20500 0.5593 0.7021
0.5505 2.2269 20750 0.5608 0.7029
0.5424 2.2537 21000 0.5644 0.7021
0.5405 2.2805 21250 0.5607 0.7013
0.5492 2.3074 21500 0.5611 0.6984
0.5589 2.3342 21750 0.5621 0.6961
0.5531 2.3610 22000 0.5615 0.6995
0.5539 2.3879 22250 0.5623 0.6950
0.5479 2.4147 22500 0.5615 0.7021
0.5476 2.4415 22750 0.5600 0.7015
0.5589 2.4683 23000 0.5596 0.6981
0.5511 2.4952 23250 0.5603 0.6997
0.5517 2.5220 23500 0.5594 0.7015
0.5439 2.5488 23750 0.5623 0.6947
0.5442 2.5757 24000 0.5612 0.7044
0.5455 2.6025 24250 0.5596 0.6966
0.5525 2.6293 24500 0.5613 0.6981
0.5384 2.6561 24750 0.5622 0.7010
0.552 2.6830 25000 0.5611 0.6981
0.5551 2.7098 25250 0.5642 0.6940
0.5411 2.7366 25500 0.5615 0.7005
0.5661 2.7635 25750 0.5614 0.6979
0.5528 2.7903 26000 0.5593 0.7002
0.5603 2.8171 26250 0.5588 0.7002
0.5514 2.8440 26500 0.5590 0.7000
0.5559 2.8708 26750 0.5591 0.7010
0.5587 2.8976 27000 0.5597 0.6997
0.5368 2.9244 27250 0.5597 0.7008
0.5624 2.9513 27500 0.5592 0.7008
0.571 2.9781 27750 0.5590 0.7005

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.0.1+cu118
  • Datasets 2.19.2
  • Tokenizers 0.19.1