Edit model card

mt5-small-task3-dataset4

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6024
  • Accuracy: 0.038
  • Mse: 6.4524
  • Log-distance: 0.6628
  • S Score: 0.4912

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: 5.6e-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: linear
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss Accuracy Mse Log-distance S Score
12.4369 1.0 250 2.2550 0.038 6.7116 0.6817 0.4748
3.1387 2.0 500 1.8576 0.024 5.8813 0.7455 0.4552
2.3265 3.0 750 1.6663 0.05 7.7168 0.7232 0.4556
1.9955 4.0 1000 1.6121 0.04 6.3175 0.6614 0.4908
1.8234 5.0 1250 1.6380 0.034 6.8099 0.6780 0.4860
1.7591 6.0 1500 1.5953 0.04 6.3175 0.6614 0.4908
1.7194 7.0 1750 1.5996 0.054 6.0821 0.6559 0.4976
1.6988 8.0 2000 1.5970 0.048 6.6575 0.6618 0.4952
1.6831 9.0 2250 1.6024 0.038 6.4524 0.6628 0.4912

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0
Downloads last month
0
Safetensors
Model size
300M params
Tensor type
F32
·

Finetuned from