Version_weird_ASAP_FineTuningBERT_AugV12_k4_task1_organization_k4_k4_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: 0.5833
• Qwk: 0.6360
• Mse: 0.5835
• Rmse: 0.7639

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	12.5034	-0.0020	12.5002	3.5356
No log	2.0	4	8.3897	0.0037	8.3869	2.8960
No log	3.0	6	6.3653	0.0	6.3630	2.5225
No log	4.0	8	6.1554	0.0	6.1531	2.4805
No log	5.0	10	6.0165	0.0	6.0142	2.4524
No log	6.0	12	5.8634	0.0	5.8612	2.4210
No log	7.0	14	5.6874	-0.0116	5.6852	2.3844
No log	8.0	16	5.4621	0.0	5.4600	2.3367
No log	9.0	18	5.0674	0.0	5.0653	2.2506
No log	10.0	20	4.4759	0.0	4.4739	2.1152
No log	11.0	22	3.8115	0.0	3.8096	1.9518
No log	12.0	24	3.3092	0.0	3.3073	1.8186
No log	13.0	26	2.8915	0.0	2.8897	1.6999
No log	14.0	28	2.3926	-0.0052	2.3909	1.5462
No log	15.0	30	1.8879	0.0775	1.8863	1.3734
No log	16.0	32	1.5242	0.0443	1.5227	1.2340
No log	17.0	34	1.3924	0.0443	1.3909	1.1794
No log	18.0	36	1.1098	0.0443	1.1085	1.0528
No log	19.0	38	0.8987	0.0427	0.8975	0.9474
No log	20.0	40	0.8200	0.3204	0.8189	0.9049
No log	21.0	42	0.7605	0.3571	0.7595	0.8715
No log	22.0	44	0.7864	0.2386	0.7857	0.8864
No log	23.0	46	0.6612	0.4274	0.6603	0.8126
No log	24.0	48	0.6539	0.3900	0.6534	0.8083
No log	25.0	50	0.5958	0.5034	0.5951	0.7714
No log	26.0	52	0.5802	0.5228	0.5796	0.7613
No log	27.0	54	0.6060	0.5756	0.6058	0.7783
No log	28.0	56	0.7729	0.4983	0.7718	0.8785
No log	29.0	58	0.8793	0.5342	0.8795	0.9378
No log	30.0	60	0.7162	0.5870	0.7162	0.8463
No log	31.0	62	0.6815	0.6182	0.6816	0.8256
No log	32.0	64	1.0220	0.5485	1.0224	1.0112
No log	33.0	66	0.8476	0.5967	0.8480	0.9209
No log	34.0	68	0.8040	0.6256	0.8045	0.8969
No log	35.0	70	0.8548	0.6130	0.8554	0.9249
No log	36.0	72	0.8360	0.6097	0.8366	0.9146
No log	37.0	74	0.5906	0.6741	0.5908	0.7687
No log	38.0	76	1.1955	0.5262	1.1965	1.0938
No log	39.0	78	0.7989	0.6282	0.7996	0.8942
No log	40.0	80	0.7281	0.5577	0.7275	0.8530
No log	41.0	82	0.7894	0.6188	0.7901	0.8889
No log	42.0	84	1.5941	0.4605	1.5951	1.2630
No log	43.0	86	0.6386	0.6294	0.6390	0.7994
No log	44.0	88	0.7303	0.6086	0.7307	0.8548
No log	45.0	90	2.4491	0.3427	2.4500	1.5652
No log	46.0	92	2.6163	0.3282	2.6172	1.6178
No log	47.0	94	0.8246	0.6022	0.8254	0.9085
No log	48.0	96	0.6979	0.6280	0.6985	0.8358
No log	49.0	98	1.3443	0.4983	1.3454	1.1599
No log	50.0	100	0.8539	0.5902	0.8547	0.9245
No log	51.0	102	0.7015	0.6580	0.7021	0.8379
No log	52.0	104	0.9372	0.5923	0.9381	0.9686
No log	53.0	106	0.6184	0.6314	0.6188	0.7866
No log	54.0	108	0.6768	0.6571	0.6775	0.8231
No log	55.0	110	1.3143	0.5009	1.3156	1.1470
No log	56.0	112	0.9127	0.5827	0.9137	0.9559
No log	57.0	114	0.5833	0.6360	0.5835	0.7639

Framework versions

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