Version_weird_ASAP_FineTuningBERT_AugV12_k4_task1_organization_k4_k4_fold4

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: 0.9100
• Qwk: 0.6018
• Mse: 0.9100
• Rmse: 0.9539

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	2	9.8699	0.0018	9.8699	3.1416
No log	2.0	4	9.2603	0.0018	9.2603	3.0431
No log	3.0	6	8.2500	0.0	8.2500	2.8723
No log	4.0	8	6.9629	0.0	6.9629	2.6387
No log	5.0	10	5.5012	0.0539	5.5012	2.3455
No log	6.0	12	4.3775	0.0118	4.3775	2.0923
No log	7.0	14	3.5394	0.0079	3.5394	1.8813
No log	8.0	16	3.0504	0.0079	3.0504	1.7465
No log	9.0	18	2.4227	0.0695	2.4227	1.5565
No log	10.0	20	1.9807	0.0527	1.9807	1.4074
No log	11.0	22	1.4979	0.0420	1.4979	1.2239
No log	12.0	24	1.2585	0.0316	1.2585	1.1218
No log	13.0	26	0.9940	0.0213	0.9940	0.9970
No log	14.0	28	0.9042	0.1417	0.9042	0.9509
No log	15.0	30	0.8174	0.1279	0.8174	0.9041
No log	16.0	32	0.8054	0.1037	0.8054	0.8974
No log	17.0	34	0.8457	0.1037	0.8457	0.9196
No log	18.0	36	0.8821	0.1180	0.8821	0.9392
No log	19.0	38	0.8497	0.1697	0.8497	0.9218
No log	20.0	40	0.7589	0.2825	0.7589	0.8712
No log	21.0	42	0.7372	0.4484	0.7372	0.8586
No log	22.0	44	0.7414	0.4687	0.7414	0.8611
No log	23.0	46	0.6761	0.5488	0.6761	0.8223
No log	24.0	48	0.7104	0.5783	0.7104	0.8428
No log	25.0	50	0.6608	0.6065	0.6608	0.8129
No log	26.0	52	0.6268	0.6399	0.6268	0.7917
No log	27.0	54	0.7651	0.6318	0.7651	0.8747
No log	28.0	56	0.7345	0.6432	0.7345	0.8570
No log	29.0	58	0.6917	0.6688	0.6917	0.8317
No log	30.0	60	0.7870	0.6525	0.7870	0.8871
No log	31.0	62	0.8858	0.6258	0.8858	0.9412
No log	32.0	64	0.8205	0.6406	0.8205	0.9058
No log	33.0	66	0.9074	0.6004	0.9074	0.9526
No log	34.0	68	0.5528	0.6653	0.5528	0.7435
No log	35.0	70	0.7237	0.6335	0.7237	0.8507
No log	36.0	72	1.0646	0.5626	1.0646	1.0318
No log	37.0	74	0.7805	0.6131	0.7805	0.8834
No log	38.0	76	0.9854	0.5777	0.9854	0.9927
No log	39.0	78	0.7791	0.6156	0.7791	0.8826
No log	40.0	80	0.8045	0.6230	0.8045	0.8969
No log	41.0	82	0.8776	0.6056	0.8776	0.9368
No log	42.0	84	0.7172	0.6423	0.7172	0.8469
No log	43.0	86	0.8559	0.6082	0.8559	0.9252
No log	44.0	88	0.8668	0.6125	0.8668	0.9310
No log	45.0	90	0.6611	0.6768	0.6611	0.8131
No log	46.0	92	0.6978	0.6660	0.6978	0.8353
No log	47.0	94	0.8708	0.6268	0.8708	0.9331
No log	48.0	96	0.8408	0.6226	0.8408	0.9169
No log	49.0	98	0.6512	0.6590	0.6512	0.8069
No log	50.0	100	0.6458	0.6530	0.6458	0.8036
No log	51.0	102	0.8325	0.5998	0.8325	0.9124
No log	52.0	104	0.7215	0.6278	0.7215	0.8494
No log	53.0	106	0.6613	0.6510	0.6613	0.8132
No log	54.0	108	0.9449	0.5932	0.9449	0.9721
No log	55.0	110	1.0646	0.5841	1.0646	1.0318
No log	56.0	112	0.8287	0.6193	0.8287	0.9103
No log	57.0	114	0.8628	0.6048	0.8628	0.9289
No log	58.0	116	0.9665	0.5865	0.9665	0.9831
No log	59.0	118	0.8327	0.6197	0.8327	0.9125
No log	60.0	120	0.7712	0.6190	0.7712	0.8782
No log	61.0	122	0.7282	0.6376	0.7282	0.8533
No log	62.0	124	0.8359	0.6182	0.8359	0.9143
No log	63.0	126	0.8933	0.5997	0.8933	0.9451
No log	64.0	128	0.9936	0.5887	0.9936	0.9968
No log	65.0	130	0.9100	0.6018	0.9100	0.9539

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0