Edit model card

TCS_Pair

This model is a fine-tuned version of mgh6/esm_cluster_35M_linear on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1515

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: 0.001
  • train_batch_size: 80
  • eval_batch_size: 80
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 8

Training results

Training Loss Epoch Step Validation Loss
0.1694 0.13 100 0.1876
0.1104 0.27 200 0.1856
0.0976 0.4 300 0.2082
0.0858 0.54 400 0.1908
0.0832 0.67 500 0.1575
0.0853 0.81 600 0.1549
0.0788 0.94 700 0.1598
0.0629 1.07 800 0.1386
0.0579 1.21 900 0.2071
0.0571 1.34 1000 0.1667
0.0567 1.48 1100 0.1421
0.0527 1.61 1200 0.1623
0.0553 1.74 1300 0.1827
0.0557 1.88 1400 0.1535
0.0523 2.01 1500 0.1353
0.0358 2.15 1600 0.1409
0.041 2.28 1700 0.1502
0.042 2.42 1800 0.1427
0.04 2.55 1900 0.1505
0.04 2.68 2000 0.1366
0.0392 2.82 2100 0.1236
0.0409 2.95 2200 0.1210
0.0327 3.09 2300 0.1448
0.0293 3.22 2400 0.1535
0.0265 3.36 2500 0.1529
0.0268 3.49 2600 0.1557
0.0241 3.62 2700 0.1505
0.0264 3.76 2800 0.1447
0.0286 3.89 2900 0.1442
0.0306 4.03 3000 0.1504
0.0152 4.16 3100 0.1311
0.0185 4.3 3200 0.1873
0.0212 4.43 3300 0.1396
0.022 4.56 3400 0.1663
0.0206 4.7 3500 0.1443
0.0231 4.83 3600 0.1483
0.0172 4.97 3700 0.1413
0.0134 5.1 3800 0.1472
0.0129 5.23 3900 0.1442
0.0126 5.37 4000 0.1369
0.013 5.5 4100 0.1576
0.012 5.64 4200 0.1348
0.013 5.77 4300 0.1572
0.0123 5.91 4400 0.1529
0.0099 6.04 4500 0.1425
0.0072 6.17 4600 0.1540
0.0081 6.31 4700 0.1463
0.0075 6.44 4800 0.1540
0.0078 6.58 4900 0.1418
0.0062 6.71 5000 0.1640
0.0062 6.85 5100 0.1392
0.0068 6.98 5200 0.1572
0.0054 7.11 5300 0.1485
0.0046 7.25 5400 0.1445
0.0054 7.38 5500 0.1488
0.0033 7.52 5600 0.1478
0.0041 7.65 5700 0.1514
0.0044 7.79 5800 0.1508
0.0025 7.92 5900 0.1515

Framework versions

  • Transformers 4.30.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.13.1
  • Tokenizers 0.13.3
Downloads last month
46
Safetensors
Model size
34M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.