Edit model card

existential-there-quantifier-transformer-4

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

  • Loss: 3.8637

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: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 4
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 3052726

Training results

Training Loss Epoch Step Validation Loss
4.2235 0.03 76320 4.1958
4.0188 1.03 152640 4.0280
3.91 0.03 228960 3.9539
3.842 1.03 305280 3.9126
3.7897 0.03 381600 3.8869
3.7491 1.03 457920 3.8716
3.7159 0.03 534240 3.8599
3.6834 1.03 610560 3.8530
3.6553 0.03 686880 3.8482
3.628 1.03 763200 3.8453
3.605 0.03 839520 3.8447
3.5866 1.03 915840 3.8442
3.57 0.03 992160 3.8431
3.5489 1.03 1068480 3.8447
3.5349 0.03 1144800 3.8466
3.5248 1.03 1221120 3.8464
3.5096 0.03 1297440 3.8480
3.4935 1.03 1373760 3.8504
3.4796 0.03 1450080 3.8505
3.4725 1.03 1526400 3.8529
3.4618 0.03 1602720 3.8541
3.4538 1.03 1679040 3.8553
3.4437 0.03 1755360 3.8561
3.433 1.03 1831680 3.8574
3.4159 0.03 1908000 3.8589
3.4048 1.03 1984320 3.8615
3.3929 0.03 2060640 3.8618
3.3857 1.03 2136960 3.8629
3.3765 0.03 2213280 3.8634
3.3637 0.03 2289600 3.8657
3.3528 0.03 2365920 3.8668
3.3489 1.03 2442240 3.8667
3.338 0.03 2518560 3.8668
3.3283 1.03 2594880 3.8668
3.3179 0.03 2671200 3.8676
3.3121 1.03 2747520 3.8667
3.3055 0.03 2823840 3.8658
3.2992 0.03 2900160 3.8658
3.2958 1.03 2976480 3.8648
3.2866 0.02 3052726 3.8637

Framework versions

  • Transformers 4.33.3
  • Pytorch 2.0.1
  • Datasets 2.12.0
  • Tokenizers 0.13.3
Downloads last month
20
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.