This model is a fine-tuned version of Hartunka/distilbert_rand_50_v2 on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

Loss: 0.5833
Accuracy: 0.7108
F1: 0.8156
Combined Score: 0.7632

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	F1	Combined Score
0.6343	1.0	15	0.6178	0.6691	0.7769	0.7230
0.5869	2.0	30	0.5833	0.7108	0.8156	0.7632
0.5196	3.0	45	0.5840	0.7108	0.7958	0.7533
0.4151	4.0	60	0.6888	0.6814	0.7797	0.7305
0.2688	5.0	75	0.9317	0.6765	0.7471	0.7118
0.1493	6.0	90	1.0192	0.7034	0.7888	0.7461
0.1019	7.0	105	1.2262	0.6936	0.7788	0.7362

Framework versions

Transformers 4.50.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.21.1

Hartunka/distilbert_rand_50_v2_mrpc

distilbert_rand_50_v2_mrpc

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Hartunka/
distilbert_rand_50_v2_mrpc