gpt-medmentions / README.md
Ben10x's picture
Model save
5e0d0b5 verified
|
raw
history blame
2.15 kB
metadata
library_name: transformers
license: mit
base_model: EleutherAI/gpt-neo-1.3B
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: gpt-medmentions
    results: []

gpt-medmentions

This model is a fine-tuned version of EleutherAI/gpt-neo-1.3B on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.4512
  • Precision: 0.4110
  • Recall: 0.5052
  • F1: 0.4533
  • Accuracy: 0.4334

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: 4
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 2
  • total_train_batch_size: 8
  • total_eval_batch_size: 16
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 5.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
1.2514 1.0 2917 1.2192 0.4016 0.3853 0.3933 0.4191
1.0825 2.0 5834 1.1892 0.4234 0.6373 0.5088 0.4324
0.9093 3.0 8751 1.2419 0.4105 0.4848 0.4446 0.4308
0.7938 4.0 11668 1.3655 0.3774 0.3369 0.3560 0.4301
0.746 5.0 14585 1.4512 0.4110 0.5052 0.4533 0.4334

Framework versions

  • Transformers 4.50.3
  • Pytorch 2.6.0+cu124
  • Datasets 3.5.0
  • Tokenizers 0.21.1