results_ft

This model is a fine-tuned version of cointegrated/rubert-tiny2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.8178

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: 256
  • eval_batch_size: 16
  • 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: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss
5.6015 0.2410 200 4.9526
4.8813 0.4819 400 4.6280
4.6572 0.7229 600 4.4362
4.5263 0.9639 800 4.3084
4.4158 1.2048 1000 4.2763
4.3402 1.4458 1200 4.1772
4.2778 1.6867 1400 4.1290
4.201 1.9277 1600 4.0205
4.1524 2.1687 1800 3.8778
4.1037 2.4096 2000 3.9216
4.0602 2.6506 2200 3.8826
4.0368 2.8916 2400 3.8800
3.9979 3.1325 2600 3.8768
3.9838 3.3735 2800 3.8561
3.9508 3.6145 3000 3.7986
3.9362 3.8554 3200 3.8220
3.92 4.0964 3400 3.7756
3.918 4.3373 3600 3.8481
3.8876 4.5783 3800 3.7696
3.8924 4.8193 4000 3.7700

Framework versions

  • Transformers 4.52.4
  • Pytorch 2.6.0+cu124
  • Datasets 3.6.0
  • Tokenizers 0.21.2
Downloads last month
10
Safetensors
Model size
29.3M params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for Dersty/rubert_tiny2_X5_ner_MLM

Finetuned
(57)
this model
Finetunes
2 models