README.md

metadata

library_name: transformers
license: mit
base_model: prajjwal1/bert-tiny
tags:
  - generated_from_trainer
metrics:
  - f1
model-index:
  - name: test
    results: []

test

This model is a fine-tuned version of prajjwal1/bert-tiny on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6897
• F1: 0.0
• Roc Auc: 0.5

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc
No log	1.0	1	0.6870	0.0	0.5
No log	2.0	2	0.6871	0.0	0.5
No log	3.0	3	0.6873	0.0	0.5
No log	4.0	4	0.6876	0.0	0.5
No log	5.0	5	0.6879	0.0	0.5
No log	6.0	6	0.6887	0.0	0.5
No log	7.0	7	0.6893	0.0	0.5
No log	8.0	8	0.6891	0.0	0.5
No log	9.0	9	0.6893	0.0	0.5
No log	10.0	10	0.6889	0.0	0.5
No log	11.0	11	0.6889	0.0	0.5
No log	12.0	12	0.6885	0.0	0.5
No log	13.0	13	0.6881	0.0	0.5
No log	14.0	14	0.6879	0.0	0.5
No log	15.0	15	0.6876	0.0	0.5
No log	16.0	16	0.6874	0.0	0.5
No log	17.0	17	0.6871	0.0	0.5
No log	18.0	18	0.6869	0.0	0.5
No log	19.0	19	0.6868	0.0	0.5
No log	20.0	20	0.6867	0.0	0.5

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cpu
• Datasets 4.1.0
• Tokenizers 0.22.0