Edit model card

xlm-roberta-base_lr0.0001_seed42_basic_original_esp-kin-eng_train

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

  • Loss: 0.0394
  • Spearman Corr: nan

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: 0.0001
  • train_batch_size: 32
  • eval_batch_size: 128
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Spearman Corr
No log 1.63 200 0.0478 0.0585
0.059 3.27 400 0.0393 0.0110
0.0534 4.9 600 0.0413 nan
0.0524 6.53 800 0.0384 nan
0.0517 8.16 1000 0.0424 nan
0.0517 9.8 1200 0.0384 nan
0.0509 11.43 1400 0.0467 nan
0.0511 13.06 1600 0.0404 nan
0.0505 14.69 1800 0.0392 -0.0320
0.05 16.33 2000 0.0382 -0.0245
0.05 17.96 2200 0.0395 nan
0.0498 19.59 2400 0.0389 -0.0672
0.0498 21.22 2600 0.0384 -0.0760
0.05 22.86 2800 0.0399 nan
0.0492 24.49 3000 0.0387 nan
0.0493 26.12 3200 0.0385 nan
0.0493 27.76 3400 0.0392 nan
0.0491 29.39 3600 0.0394 nan

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.2
Downloads last month
2
Safetensors
Model size
278M params
Tensor type
F32
·

Finetuned from