Edit model card

fresh-2-layer-medmcqa10000-distill-of-fresh-2-layer-gpqa_EVAL_gpqa

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

  • Loss: 7.2371
  • Accuracy: 0.7222

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: 0.0005
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 321
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 5000

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 0.32 100 13.8640 0.4495
No log 0.64 200 11.8895 0.5152
No log 0.96 300 10.3828 0.5303
No log 1.28 400 9.0300 0.5657
3.3496 1.6 500 8.6264 0.5808
3.3496 1.92 600 8.8207 0.6061
3.3496 2.24 700 8.8302 0.5960
3.3496 2.56 800 8.8428 0.6566
3.3496 2.88 900 7.9552 0.6364
0.8968 3.19 1000 8.4617 0.6263
0.8968 3.51 1100 8.8555 0.6616
0.8968 3.83 1200 7.5445 0.6566
0.8968 4.15 1300 7.6791 0.6717
0.8968 4.47 1400 7.8363 0.6616
0.4853 4.79 1500 7.6269 0.6515
0.4853 5.11 1600 7.5024 0.6919
0.4853 5.43 1700 7.4191 0.6717
0.4853 5.75 1800 7.6877 0.6768
0.4853 6.07 1900 7.4651 0.6818
0.3197 6.39 2000 7.4452 0.6970
0.3197 6.71 2100 7.2401 0.7121
0.3197 7.03 2200 7.4038 0.7121
0.3197 7.35 2300 7.1982 0.7071
0.3197 7.67 2400 7.2287 0.7071
0.2394 7.99 2500 7.2371 0.7222
0.2394 8.31 2600 7.2513 0.7071
0.2394 8.63 2700 7.3788 0.6919
0.2394 8.95 2800 7.1303 0.7071
0.2394 9.27 2900 7.1608 0.7121
0.1744 9.58 3000 7.1039 0.7222

Framework versions

  • Transformers 4.34.0.dev0
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.5
  • Tokenizers 0.14.0
Downloads last month
4
Inference API (serverless) does not yet support transformers models for this pipeline type.