Version12LowDevASAP_FineTuningBERT_AugV12_k3_task1_organization_k3_k3_fold3

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5465
• Qwk: 0.6224
• Mse: 0.5465
• Rmse: 0.7392

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: 2e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	1	13.2112	-0.0041	13.2112	3.6347
No log	2.0	2	11.8586	-0.0120	11.8586	3.4436
No log	3.0	3	10.4113	0.0	10.4113	3.2267
No log	4.0	4	9.3190	0.0	9.3190	3.0527
No log	5.0	5	8.3021	0.0	8.3021	2.8813
No log	6.0	6	7.4148	0.0	7.4148	2.7230
No log	7.0	7	6.7478	0.0	6.7478	2.5977
No log	8.0	8	5.9926	0.0113	5.9926	2.4480
No log	9.0	9	5.2420	0.0	5.2420	2.2895
No log	10.0	10	4.6642	0.0	4.6642	2.1597
No log	11.0	11	4.1083	0.0	4.1083	2.0269
No log	12.0	12	3.6905	0.0	3.6905	1.9211
No log	13.0	13	3.2561	0.0	3.2561	1.8045
No log	14.0	14	2.8932	0.0	2.8932	1.7009
No log	15.0	15	2.5906	0.1709	2.5906	1.6095
No log	16.0	16	2.3207	0.0132	2.3207	1.5234
No log	17.0	17	2.0782	0.0	2.0782	1.4416
No log	18.0	18	1.8751	0.0	1.8751	1.3693
No log	19.0	19	1.7087	0.0	1.7087	1.3072
No log	20.0	20	1.5467	0.0	1.5467	1.2437
No log	21.0	21	1.3859	0.0	1.3859	1.1773
No log	22.0	22	1.2598	0.0	1.2598	1.1224
No log	23.0	23	1.1578	0.0	1.1578	1.0760
No log	24.0	24	1.0654	0.0	1.0654	1.0322
No log	25.0	25	0.9921	0.0	0.9921	0.9960
No log	26.0	26	0.9458	0.1852	0.9458	0.9725
No log	27.0	27	0.8734	0.2792	0.8734	0.9345
No log	28.0	28	0.8245	0.2183	0.8245	0.9080
No log	29.0	29	0.7923	0.1973	0.7923	0.8901
No log	30.0	30	0.7685	0.1262	0.7685	0.8766
No log	31.0	31	0.7530	0.0523	0.7530	0.8678
No log	32.0	32	0.7196	0.1262	0.7196	0.8483
No log	33.0	33	0.6675	0.2453	0.6675	0.8170
No log	34.0	34	0.6590	0.2245	0.6590	0.8118
No log	35.0	35	0.6639	0.2033	0.6639	0.8148
No log	36.0	36	0.6292	0.2453	0.6292	0.7932
No log	37.0	37	0.5841	0.2857	0.5841	0.7643
No log	38.0	38	0.5431	0.2987	0.5431	0.7369
No log	39.0	39	0.5175	0.3112	0.5175	0.7194
No log	40.0	40	0.5308	0.2987	0.5308	0.7286
No log	41.0	41	0.4985	0.4038	0.4985	0.7060
No log	42.0	42	0.4735	0.5064	0.4735	0.6881
No log	43.0	43	0.4767	0.5924	0.4767	0.6904
No log	44.0	44	0.5478	0.5584	0.5478	0.7401
No log	45.0	45	0.5050	0.6222	0.5050	0.7106
No log	46.0	46	0.4641	0.6479	0.4641	0.6812
No log	47.0	47	0.4705	0.6420	0.4705	0.6860
No log	48.0	48	0.5205	0.5869	0.5205	0.7215
No log	49.0	49	0.5146	0.5850	0.5146	0.7173
No log	50.0	50	0.4845	0.6642	0.4845	0.6961
No log	51.0	51	0.4755	0.6443	0.4755	0.6895
No log	52.0	52	0.5114	0.6204	0.5114	0.7151
No log	53.0	53	0.5512	0.5988	0.5512	0.7424
No log	54.0	54	0.5970	0.5764	0.5970	0.7727
No log	55.0	55	0.5706	0.5794	0.5706	0.7554
No log	56.0	56	0.5343	0.6279	0.5343	0.7310
No log	57.0	57	0.5485	0.6319	0.5485	0.7406
No log	58.0	58	0.6027	0.5666	0.6027	0.7763
No log	59.0	59	0.6717	0.5678	0.6717	0.8196
No log	60.0	60	0.5963	0.5666	0.5963	0.7722
No log	61.0	61	0.5275	0.6139	0.5275	0.7263
No log	62.0	62	0.5147	0.6476	0.5147	0.7174
No log	63.0	63	0.5441	0.5766	0.5441	0.7376
No log	64.0	64	0.6643	0.5633	0.6643	0.8150
No log	65.0	65	0.6435	0.5803	0.6435	0.8022
No log	66.0	66	0.5400	0.5890	0.5400	0.7348
No log	67.0	67	0.5224	0.6336	0.5224	0.7228
No log	68.0	68	0.5123	0.6467	0.5123	0.7157
No log	69.0	69	0.5255	0.6358	0.5255	0.7249
No log	70.0	70	0.5465	0.6224	0.5465	0.7392

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0