tinybert_train

This model is a fine-tuned version of distilbert-base-uncased on the gokulsrinivasagan/processed_wikitext-103-raw-v1-ld dataset. It achieves the following results on the evaluation set:

Loss: 5.7861
Accuracy: 0.1607

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: 0.0001
train_batch_size: 160
eval_batch_size: 160
seed: 10
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 10000
num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
6.122	6.9979	10000	6.0554	0.1535
5.9055	13.9958	20000	5.8711	0.1556
5.8328	20.9937	30000	5.8420	0.1547

Framework versions

Transformers 4.46.1
Pytorch 2.2.0+cu121
Datasets 3.1.0
Tokenizers 0.20.1

Downloads last month: 5

Safetensors

Model size

33M params

Tensor type

F32

Model tree for gokulsrinivasagan/tinybert_train

Base model

distilbert/distilbert-base-uncased

Finetuned

(12297)

this model

Finetunes

9 models

Dataset used to train gokulsrinivasagan/tinybert_train

Evaluation results

Accuracy on gokulsrinivasagan/processed_wikitext-103-raw-v1-ld
self-reported

0.161