metadata
language:
- en
base_model: Hartunka/tiny_bert_km_100_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
- f1
model-index:
- name: tiny_bert_km_100_v1_mrpc
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE MRPC
type: glue
args: mrpc
metrics:
- name: Accuracy
type: accuracy
value: 0.7230392156862745
- name: F1
type: f1
value: 0.8197767145135566
tiny_bert_km_100_v1_mrpc
This model is a fine-tuned version of Hartunka/tiny_bert_km_100_v1 on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.5885
- Accuracy: 0.7230
- F1: 0.8198
- Combined Score: 0.7714
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 |
|---|---|---|---|---|---|---|
| 0.6262 | 1.0 | 15 | 0.6086 | 0.6961 | 0.8063 | 0.7512 |
| 0.599 | 2.0 | 30 | 0.6000 | 0.7034 | 0.8191 | 0.7613 |
| 0.5725 | 3.0 | 45 | 0.5969 | 0.7059 | 0.8176 | 0.7618 |
| 0.5494 | 4.0 | 60 | 0.5885 | 0.7230 | 0.8198 | 0.7714 |
| 0.5073 | 5.0 | 75 | 0.6107 | 0.6863 | 0.7808 | 0.7335 |
| 0.4408 | 6.0 | 90 | 0.6365 | 0.7010 | 0.7939 | 0.7474 |
| 0.3565 | 7.0 | 105 | 0.7095 | 0.7108 | 0.8013 | 0.7561 |
| 0.2434 | 8.0 | 120 | 0.8137 | 0.7034 | 0.7939 | 0.7486 |
| 0.1755 | 9.0 | 135 | 0.9716 | 0.6593 | 0.7440 | 0.7017 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.19.1