metadata
language:
- en
base_model: Hartunka/tiny_bert_rand_100_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
- f1
model-index:
- name: tiny_bert_rand_100_v1_qqp
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE QQP
type: glue
args: qqp
metrics:
- name: Accuracy
type: accuracy
value: 0.8110066782092505
- name: F1
type: f1
value: 0.7419714314659103
tiny_bert_rand_100_v1_qqp
This model is a fine-tuned version of Hartunka/tiny_bert_rand_100_v1 on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.4153
- Accuracy: 0.8110
- F1: 0.7420
- Combined Score: 0.7765
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.496 | 1.0 | 1422 | 0.4492 | 0.7838 | 0.6806 | 0.7322 |
| 0.4089 | 2.0 | 2844 | 0.4196 | 0.8034 | 0.7191 | 0.7612 |
| 0.3518 | 3.0 | 4266 | 0.4153 | 0.8110 | 0.7420 | 0.7765 |
| 0.3062 | 4.0 | 5688 | 0.4296 | 0.8186 | 0.7423 | 0.7805 |
| 0.2689 | 5.0 | 7110 | 0.4466 | 0.8198 | 0.7533 | 0.7866 |
| 0.239 | 6.0 | 8532 | 0.4361 | 0.8202 | 0.7623 | 0.7912 |
| 0.2131 | 7.0 | 9954 | 0.4664 | 0.8231 | 0.7650 | 0.7941 |
| 0.1896 | 8.0 | 11376 | 0.5052 | 0.8201 | 0.7678 | 0.7939 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.19.1