bert_base_train_book_ent_15p_stsb

This model is a fine-tuned version of gokulsrinivasagan/bert_base_train_book_ent_15p on the GLUE STSB dataset. It achieves the following results on the evaluation set:

• Loss: 1.3250
• Pearson: 0.6548
• Spearmanr: 0.6523
• Combined Score: 0.6536

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
2.5593	1.0	23	2.5903	0.1155	0.1099	0.1127
1.7953	2.0	46	1.9161	0.4393	0.4422	0.4407
1.2931	3.0	69	2.1854	0.5295	0.5599	0.5447
0.7781	4.0	92	1.6846	0.6157	0.6245	0.6201
0.575	5.0	115	1.7262	0.6215	0.6358	0.6287
0.4165	6.0	138	1.6485	0.6352	0.6420	0.6386
0.3111	7.0	161	1.4595	0.6564	0.6575	0.6569
0.2402	8.0	184	1.5228	0.6213	0.6165	0.6189
0.2177	9.0	207	1.3687	0.6513	0.6511	0.6512
0.1798	10.0	230	1.3785	0.6576	0.6535	0.6556
0.1636	11.0	253	1.3667	0.6600	0.6552	0.6576
0.1489	12.0	276	1.3250	0.6548	0.6523	0.6536
0.1319	13.0	299	1.4696	0.6477	0.6445	0.6461
0.1129	14.0	322	1.4006	0.6596	0.6575	0.6586
0.1075	15.0	345	1.4180	0.6725	0.6758	0.6741
0.1062	16.0	368	1.4378	0.6550	0.6515	0.6533
0.0981	17.0	391	1.5027	0.6609	0.6587	0.6598

Framework versions

• Transformers 4.46.3
• Pytorch 2.2.1+cu118
• Datasets 3.1.0
• Tokenizers 0.20.3