---
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_km_100_v2
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
model-index:
- name: tiny_bert_km_100_v2_qnli
  results:
  - task:
      name: Text Classification
      type: text-classification
    dataset:
      name: GLUE QNLI
      type: glue
      args: qnli
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.6112026359143328
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# tiny_bert_km_100_v2_qnli

This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v2](https://huggingface.co/Hartunka/tiny_bert_km_100_v2) on the GLUE QNLI dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6510
- Accuracy: 0.6112

## 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: 256
- eval_batch_size: 256
- seed: 10
- optimizer: Use 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: 50

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.6693        | 1.0   | 410  | 0.6510          | 0.6112   |
| 0.6414        | 2.0   | 820  | 0.6562          | 0.6156   |
| 0.6031        | 3.0   | 1230 | 0.6522          | 0.6205   |
| 0.5387        | 4.0   | 1640 | 0.6984          | 0.6108   |
| 0.4639        | 5.0   | 2050 | 0.7590          | 0.6138   |
| 0.3926        | 6.0   | 2460 | 0.9158          | 0.6039   |


### Framework versions

- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1