This model is a fine-tuned version of Hartunka/distilbert_rand_100_v2 on the GLUE STSB dataset. It achieves the following results on the evaluation set:

Loss: 2.3172
Pearson: 0.2906
Spearmanr: 0.2841
Combined Score: 0.2874

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
2.7709	1.0	23	2.5323	0.1404	0.1165	0.1284
1.9319	2.0	46	2.3231	0.1868	0.1617	0.1742
1.6901	3.0	69	2.5205	0.2196	0.2097	0.2146
1.4033	4.0	92	2.3172	0.2906	0.2841	0.2874
1.0661	5.0	115	2.4881	0.2764	0.2746	0.2755
0.8233	6.0	138	2.4506	0.3105	0.3055	0.3080
0.6527	7.0	161	2.6287	0.3095	0.3037	0.3066
0.5076	8.0	184	2.6804	0.3066	0.2993	0.3029
0.4436	9.0	207	2.5319	0.3202	0.3142	0.3172

Framework versions

Transformers 4.50.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.21.1

Hartunka/distilbert_rand_100_v2_stsb

distilbert_rand_100_v2_stsb

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Hartunka/
distilbert_rand_100_v2_stsb