Version25NewTestASAP_FineTuningBERT_AugV25_k3_task1_organization_k3_k3_fold1

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: 0.6901
• Qwk: 0.4973
• Mse: 0.6898
• Rmse: 0.8306

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: 150

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	3	12.2635	0.0	12.2610	3.5016
No log	2.0	6	9.3179	0.0	9.3154	3.0521
No log	3.0	9	5.9806	0.0	5.9784	2.4451
No log	4.0	12	4.2436	0.0	4.2415	2.0595
No log	5.0	15	2.9786	0.0	2.9767	1.7253
No log	6.0	18	1.8711	0.0315	1.8695	1.3673
No log	7.0	21	1.2265	0.0	1.2250	1.1068
No log	8.0	24	1.0083	0.0	1.0069	1.0034
No log	9.0	27	0.8872	0.1365	0.8860	0.9413
No log	10.0	30	0.9482	-0.0435	0.9470	0.9731
No log	11.0	33	1.3545	0.0878	1.3530	1.1632
No log	12.0	36	0.9426	0.1129	0.9413	0.9702
No log	13.0	39	0.6980	0.3823	0.6969	0.8348
No log	14.0	42	0.8877	0.2349	0.8867	0.9416
No log	15.0	45	0.8350	0.1433	0.8341	0.9133
No log	16.0	48	0.7642	0.2831	0.7634	0.8737
No log	17.0	51	0.6726	0.4269	0.6718	0.8196
No log	18.0	54	1.0985	0.3174	1.0975	1.0476
No log	19.0	57	0.5912	0.5232	0.5904	0.7684
No log	20.0	60	0.6026	0.4951	0.6017	0.7757
No log	21.0	63	1.3306	0.2790	1.3295	1.1530
No log	22.0	66	0.9269	0.3694	0.9260	0.9623
No log	23.0	69	0.7704	0.4059	0.7698	0.8774
No log	24.0	72	0.7499	0.3730	0.7494	0.8657
No log	25.0	75	0.7505	0.3937	0.7503	0.8662
No log	26.0	78	0.7336	0.4947	0.7334	0.8564
No log	27.0	81	0.7949	0.4595	0.7947	0.8914
No log	28.0	84	0.6688	0.5679	0.6687	0.8178
No log	29.0	87	1.0186	0.4395	1.0184	1.0092
No log	30.0	90	0.7249	0.5245	0.7248	0.8514
No log	31.0	93	0.7005	0.5125	0.7004	0.8369
No log	32.0	96	0.6824	0.5314	0.6822	0.8260
No log	33.0	99	0.9515	0.4746	0.9515	0.9755
No log	34.0	102	0.7790	0.4878	0.7791	0.8827
No log	35.0	105	0.7954	0.5170	0.7954	0.8919
No log	36.0	108	0.7121	0.5078	0.7120	0.8438
No log	37.0	111	0.7370	0.4720	0.7369	0.8585
No log	38.0	114	0.8485	0.4879	0.8485	0.9212
No log	39.0	117	0.7932	0.4812	0.7932	0.8906
No log	40.0	120	0.7177	0.4799	0.7175	0.8470
No log	41.0	123	0.7262	0.5071	0.7260	0.8521
No log	42.0	126	1.0833	0.4488	1.0832	1.0408
No log	43.0	129	0.7141	0.4741	0.7140	0.8450
No log	44.0	132	0.8622	0.4189	0.8620	0.9284
No log	45.0	135	0.7836	0.5149	0.7836	0.8852
No log	46.0	138	0.7314	0.5028	0.7313	0.8552
No log	47.0	141	0.8305	0.4828	0.8305	0.9113
No log	48.0	144	0.7711	0.4631	0.7711	0.8781
No log	49.0	147	0.7939	0.4743	0.7938	0.8909
No log	50.0	150	0.7316	0.4491	0.7314	0.8552
No log	51.0	153	0.8450	0.4926	0.8449	0.9192
No log	52.0	156	0.8706	0.4821	0.8705	0.9330
No log	53.0	159	0.7451	0.4537	0.7449	0.8630
No log	54.0	162	0.7548	0.4621	0.7546	0.8687
No log	55.0	165	0.7269	0.4826	0.7267	0.8524
No log	56.0	168	0.9016	0.4668	0.9015	0.9495
No log	57.0	171	0.8755	0.4676	0.8753	0.9356
No log	58.0	174	0.6901	0.4973	0.6898	0.8306

Framework versions

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