Version_weird_ASAP_FineTuningBERT_AugV12_k5_task1_organization_k5_k5_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: 0.7043
• Qwk: 0.6514
• Mse: 0.7043
• Rmse: 0.8392

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	7.8281	0.0	7.8281	2.7979
No log	2.0	4	7.4864	0.0	7.4864	2.7361
No log	3.0	6	7.0304	0.0	7.0304	2.6515
No log	4.0	8	6.2152	-0.0081	6.2152	2.4930
No log	5.0	10	4.8503	0.0040	4.8503	2.2023
No log	6.0	12	3.7824	0.0040	3.7824	1.9448
No log	7.0	14	2.9749	0.0040	2.9749	1.7248
No log	8.0	16	2.2869	0.1396	2.2869	1.5123
No log	9.0	18	1.7725	0.0317	1.7725	1.3313
No log	10.0	20	1.4059	0.0213	1.4059	1.1857
No log	11.0	22	1.1507	0.0107	1.1507	1.0727
No log	12.0	24	0.9888	0.0107	0.9888	0.9944
No log	13.0	26	0.8605	0.3066	0.8605	0.9276
No log	14.0	28	0.8392	0.1058	0.8392	0.9161
No log	15.0	30	0.7959	0.1763	0.7959	0.8921
No log	16.0	32	0.8009	0.1395	0.8009	0.8949
No log	17.0	34	0.8958	0.1020	0.8958	0.9465
No log	18.0	36	0.8088	0.1525	0.8088	0.8993
No log	19.0	38	0.8837	0.2660	0.8837	0.9401
No log	20.0	40	0.7428	0.3789	0.7428	0.8619
No log	21.0	42	0.6486	0.4686	0.6486	0.8054
No log	22.0	44	0.7619	0.5163	0.7619	0.8729
No log	23.0	46	0.6691	0.5392	0.6691	0.8180
No log	24.0	48	0.7095	0.5503	0.7095	0.8423
No log	25.0	50	0.7315	0.5878	0.7315	0.8553
No log	26.0	52	0.7964	0.6106	0.7964	0.8924
No log	27.0	54	0.7203	0.6260	0.7203	0.8487
No log	28.0	56	0.7946	0.6174	0.7946	0.8914
No log	29.0	58	0.5887	0.6559	0.5887	0.7672
No log	30.0	60	1.2176	0.5273	1.2176	1.1034
No log	31.0	62	1.1290	0.5425	1.1290	1.0625
No log	32.0	64	0.6051	0.6353	0.6051	0.7779
No log	33.0	66	0.7445	0.6225	0.7445	0.8629
No log	34.0	68	0.8563	0.6158	0.8563	0.9254
No log	35.0	70	0.6580	0.6315	0.6580	0.8111
No log	36.0	72	0.9026	0.5908	0.9026	0.9500
No log	37.0	74	0.9584	0.5651	0.9584	0.9790
No log	38.0	76	0.6017	0.6379	0.6017	0.7757
No log	39.0	78	0.6475	0.6380	0.6475	0.8047
No log	40.0	80	0.6570	0.6329	0.6570	0.8105
No log	41.0	82	0.5912	0.6592	0.5912	0.7689
No log	42.0	84	0.7130	0.6516	0.7130	0.8444
No log	43.0	86	0.6370	0.6577	0.6370	0.7981
No log	44.0	88	0.7314	0.6479	0.7314	0.8552
No log	45.0	90	1.0826	0.5579	1.0826	1.0405
No log	46.0	92	0.7245	0.6549	0.7245	0.8512
No log	47.0	94	0.6979	0.6624	0.6979	0.8354
No log	48.0	96	0.8069	0.6351	0.8069	0.8983
No log	49.0	98	0.5772	0.6735	0.5772	0.7597
No log	50.0	100	0.5622	0.6785	0.5622	0.7498
No log	51.0	102	0.8310	0.6071	0.8310	0.9116
No log	52.0	104	0.9788	0.5569	0.9788	0.9893
No log	53.0	106	0.6793	0.6522	0.6793	0.8242
No log	54.0	108	0.6501	0.6514	0.6501	0.8063
No log	55.0	110	0.7040	0.6443	0.7040	0.8391
No log	56.0	112	0.7175	0.6454	0.7175	0.8470
No log	57.0	114	0.6276	0.6610	0.6276	0.7922
No log	58.0	116	0.7006	0.6563	0.7006	0.8370
No log	59.0	118	0.9054	0.5961	0.9054	0.9515
No log	60.0	120	0.7531	0.6306	0.7531	0.8678
No log	61.0	122	0.7148	0.6485	0.7148	0.8455
No log	62.0	124	0.8133	0.6179	0.8133	0.9019
No log	63.0	126	0.6847	0.6551	0.6847	0.8275
No log	64.0	128	0.6755	0.6535	0.6755	0.8219
No log	65.0	130	0.7460	0.6632	0.7460	0.8637
No log	66.0	132	0.7048	0.6638	0.7048	0.8395
No log	67.0	134	0.7443	0.6461	0.7443	0.8627
No log	68.0	136	0.6997	0.6662	0.6997	0.8365
No log	69.0	138	0.7318	0.6551	0.7318	0.8555
No log	70.0	140	0.7043	0.6514	0.7043	0.8392

Framework versions

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