lukasdrg's picture
End of training
bc82c36
metadata
license: apache-2.0
base_model: allenai/longformer-base-4096
tags:
  - generated_from_trainer
model-index:
  - name: clinical_longformer_same_tokens_exp_gpu_vtres
    results: []

clinical_longformer_same_tokens_exp_gpu_vtres

This model is a fine-tuned version of allenai/longformer-base-4096 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9360

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: 2e-05
  • train_batch_size: 4
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 128
  • total_train_batch_size: 512
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1500
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss
3.4475 0.28 129 2.4612
3.2401 0.57 258 2.2069
3.0136 0.85 387 2.0084
2.5518 1.13 516 1.8632
2.9398 1.42 645 1.7412
2.0755 1.7 774 1.6260
2.231 1.98 903 1.5247
1.9556 2.27 1032 1.4482
2.0415 2.55 1161 1.3771
2.111 2.83 1290 1.3097
1.6116 3.12 1419 1.2572
1.6044 3.4 1548 1.2108
1.6614 3.68 1677 1.1802
1.5013 3.97 1806 1.1371
1.4666 4.25 1935 1.1079
1.4608 4.53 2064 1.0894
1.5144 4.82 2193 1.0672
1.3958 5.1 2322 1.0390
1.3949 5.38 2451 1.0283
1.539 5.67 2580 1.0249
1.5819 5.95 2709 1.0085
1.2588 6.23 2838 1.0033
1.2704 6.52 2967 0.9875
1.2457 6.8 3096 0.9768
1.2066 7.08 3225 0.9707
1.226 7.37 3354 0.9657
1.3292 7.65 3483 0.9590
1.1629 7.93 3612 0.9468
1.2711 8.22 3741 0.9437
1.3092 8.5 3870 0.9445
1.35 8.78 3999 0.9424
1.3402 9.07 4128 0.9360
1.1433 9.35 4257 0.9349
1.3059 9.63 4386 0.9296
1.1391 9.92 4515 0.9360

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.15.0
  • Tokenizers 0.15.0