Edit model card

Visualize in Weights & Biases

yes_no_model_dutch_6000_v2

This model is a fine-tuned version of procit001/yes_no_model_dutch_6000 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0000

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: 20
  • eval_batch_size: 20
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss
0.363 0.0476 10 0.2087
0.2209 0.0952 20 0.1218
0.1288 0.1429 30 0.0920
0.0671 0.1905 40 0.0441
0.0364 0.2381 50 0.0111
0.036 0.2857 60 0.0057
0.0278 0.3333 70 0.0035
0.0016 0.3810 80 0.0003
0.0048 0.4286 90 0.0001
0.0022 0.4762 100 0.0000
0.0011 0.5238 110 0.0000
0.0052 0.5714 120 0.0000
0.0032 0.6190 130 0.0000
0.0003 0.6667 140 0.0000
0.0088 0.7143 150 0.0000
0.0002 0.7619 160 0.0000
0.0002 0.8095 170 0.0000
0.0 0.8571 180 0.0000
0.0001 0.9048 190 0.0000
0.0 0.9524 200 0.0000
0.0001 1.0 210 0.0000
0.0001 1.0476 220 0.0000
0.0 1.0952 230 0.0000
0.0 1.1429 240 0.0000
0.0001 1.1905 250 0.0000
0.0 1.2381 260 0.0000
0.0 1.2857 270 0.0000
0.0 1.3333 280 0.0000
0.0 1.3810 290 0.0000
0.0 1.4286 300 0.0000
0.0 1.4762 310 0.0000
0.0 1.5238 320 0.0000
0.0 1.5714 330 0.0000
0.0 1.6190 340 0.0000
0.0 1.6667 350 0.0000
0.0 1.7143 360 0.0000
0.0 1.7619 370 0.0000
0.0 1.8095 380 0.0000
0.0 1.8571 390 0.0000
0.0 1.9048 400 0.0000
0.0 1.9524 410 0.0000
0.0 2.0 420 0.0000
0.0 2.0476 430 0.0000
0.0 2.0952 440 0.0000
0.0 2.1429 450 0.0000
0.0 2.1905 460 0.0000
0.0 2.2381 470 0.0000
0.0 2.2857 480 0.0000
0.0 2.3333 490 0.0000
0.0 2.3810 500 0.0000
0.0 2.4286 510 0.0000
0.0 2.4762 520 0.0000
0.0 2.5238 530 0.0000
0.0 2.5714 540 0.0000
0.0 2.6190 550 0.0000
0.0 2.6667 560 0.0000
0.0 2.7143 570 0.0000
0.0 2.7619 580 0.0000
0.0 2.8095 590 0.0000
0.0 2.8571 600 0.0000
0.0 2.9048 610 0.0000
0.0 2.9524 620 0.0000
0.0006 3.0 630 0.0000

Framework versions

  • Transformers 4.42.3
  • Pytorch 2.3.1+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1
Downloads last month
2
Safetensors
Model size
124M params
Tensor type
F32
·
Inference API
This model can be loaded on Inference API (serverless).

Finetuned from