metadata
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_rand_10_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: tiny_bert_rand_10_v1_stsb
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE STSB
type: glue
args: stsb
metrics:
- name: Spearmanr
type: spearmanr
value: 0.07340024527090903
tiny_bert_rand_10_v1_stsb
This model is a fine-tuned version of Hartunka/tiny_bert_rand_10_v1 on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.3236
- Pearson: 0.0790
- Spearmanr: 0.0734
- Combined Score: 0.0762
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 | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 2.9227 | 1.0 | 23 | 2.3236 | 0.0790 | 0.0734 | 0.0762 |
| 1.9914 | 2.0 | 46 | 2.6350 | 0.0981 | 0.0769 | 0.0875 |
| 1.8134 | 3.0 | 69 | 2.4406 | 0.1890 | 0.1829 | 0.1859 |
| 1.6045 | 4.0 | 92 | 2.4218 | 0.2333 | 0.2319 | 0.2326 |
| 1.3303 | 5.0 | 115 | 2.5452 | 0.2553 | 0.2550 | 0.2551 |
| 1.1276 | 6.0 | 138 | 2.6353 | 0.2658 | 0.2687 | 0.2673 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1