Hugging Face's logo

mtzig
/
lge_tests_prelim

Transformers
Safetensors
nanogpt
Generated from Trainer
Inference Endpoints
Model card Files Community
lge_tests_prelim / README.md
mtzig's picture
mtzig
Model save
4977831 verified
preview code
|
raw
history blame
6.32 kB
metadata
library_name: transformers
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: lge_tests_prelim
    results: []

lge_tests_prelim

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

  • Loss: 0.5307
  • Accuracy: 0.34

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: 0.0005
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 0 0 2.6404 0.0
2.5674 0.0128 100 2.5647 0.0
2.5021 0.0256 200 2.4965 0.0
2.4106 0.0384 300 2.4111 0.0
2.3432 0.0512 400 2.3422 0.0
2.32 0.0640 500 2.3004 0.0
2.2412 0.0768 600 2.2452 0.0
2.1721 0.0896 700 2.1679 0.0
1.9939 0.1024 800 1.9887 0.0
1.9089 0.1152 900 1.9041 0.0
2.0517 0.1280 1000 1.8690 0.0
1.854 0.1408 1100 1.7567 0.0
1.7972 0.1536 1200 1.7314 0.0
1.6798 0.1665 1300 1.7170 0.0
1.6579 0.1793 1400 1.6576 0.0
1.6968 0.1921 1500 1.6208 0.005
1.5677 0.2049 1600 1.6667 0.0
1.5288 0.2177 1700 1.5156 0.005
1.5954 0.2305 1800 1.5904 0.0
1.473 0.2433 1900 1.5063 0.01
1.4783 0.2561 2000 1.4800 0.01
1.5276 0.2689 2100 1.4590 0.01
1.3354 0.2817 2200 1.4401 0.02
1.4443 0.2945 2300 1.3868 0.0
1.3269 0.3073 2400 1.3720 0.025
1.3306 0.3201 2500 1.3052 0.015
1.274 0.3329 2600 1.3153 0.015
1.2331 0.3457 2700 1.2486 0.02
1.2947 0.3585 2800 1.2650 0.01
1.1635 0.3713 2900 1.1717 0.03
1.112 0.3841 3000 1.1700 0.045
1.1343 0.3969 3100 1.1362 0.04
1.072 0.4097 3200 1.1037 0.055
1.0831 0.4225 3300 1.0751 0.02
1.0762 0.4353 3400 1.0773 0.035
0.9965 0.4481 3500 1.0021 0.015
0.9867 0.4609 3600 0.9721 0.065
0.9194 0.4738 3700 0.9881 0.08
1.1577 0.4866 3800 1.1223 0.05
0.9286 0.4994 3900 0.9181 0.065
0.932 0.5122 4000 0.9695 0.035
0.907 0.5250 4100 0.9809 0.085
0.8528 0.5378 4200 0.8546 0.07
0.8456 0.5506 4300 0.8779 0.095
0.7858 0.5634 4400 0.8470 0.08
0.8417 0.5762 4500 0.8280 0.09
0.8261 0.5890 4600 0.8270 0.11
0.8291 0.6018 4700 0.8272 0.07
0.782 0.6146 4800 0.7997 0.07
0.7449 0.6274 4900 0.7533 0.06
0.7362 0.6402 5000 0.7722 0.1
0.7751 0.6530 5100 0.7441 0.11
0.7249 0.6658 5200 0.7591 0.08
0.7121 0.6786 5300 0.7160 0.17
0.704 0.6914 5400 0.7142 0.1
0.6699 0.7042 5500 0.6914 0.09
0.6853 0.7170 5600 0.6954 0.105
0.6638 0.7298 5700 0.6716 0.165
0.6862 0.7426 5800 0.6623 0.12
0.655 0.7554 5900 0.6549 0.145
0.6251 0.7682 6000 0.6537 0.125
0.637 0.7810 6100 0.6379 0.155
0.625 0.7939 6200 0.6188 0.17
0.6114 0.8067 6300 0.6036 0.205
0.6303 0.8195 6400 0.6004 0.19
0.5983 0.8323 6500 0.5845 0.225
0.6014 0.8451 6600 0.5766 0.245
0.5785 0.8579 6700 0.5765 0.24
0.5804 0.8707 6800 0.5620 0.28
0.5633 0.8835 6900 0.5518 0.3
0.5533 0.8963 7000 0.5489 0.305
0.5551 0.9091 7100 0.5481 0.305
0.569 0.9219 7200 0.5398 0.3
0.5583 0.9347 7300 0.5389 0.31
0.5357 0.9475 7400 0.5369 0.325
0.5453 0.9603 7500 0.5328 0.34
0.5472 0.9731 7600 0.5309 0.345
0.5349 0.9859 7700 0.5307 0.345
0.5309 0.9987 7800 0.5307 0.34

Framework versions

  • Transformers 4.46.0
  • Pytorch 2.5.1
  • Datasets 3.1.0
  • Tokenizers 0.20.1