Version_weird_ASAP_FineTuningBERT_AugV12_k3_task1_organization_k3_k3_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: 0.9160
• Qwk: 0.5845
• Mse: 0.9159
• Rmse: 0.9571

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	8.2939	0.0037	8.2939	2.8799
No log	2.0	2	7.3961	0.0018	7.3961	2.7196
No log	3.0	3	6.8615	0.0018	6.8615	2.6194
No log	4.0	4	6.3513	0.0018	6.3513	2.5202
No log	5.0	5	5.8530	0.0266	5.8530	2.4193
No log	6.0	6	5.3950	0.0080	5.3950	2.3227
No log	7.0	7	4.9554	0.0079	4.9554	2.2261
No log	8.0	8	4.5116	0.0040	4.5116	2.1241
No log	9.0	9	4.0895	0.0040	4.0895	2.0222
No log	10.0	10	3.6898	0.0040	3.6898	1.9209
No log	11.0	11	3.2580	0.0040	3.2580	1.8050
No log	12.0	12	2.8290	0.0040	2.8290	1.6820
No log	13.0	13	2.4312	0.0302	2.4312	1.5592
No log	14.0	14	2.1171	0.1175	2.1171	1.4550
No log	15.0	15	1.8482	0.0408	1.8482	1.3595
No log	16.0	16	1.6343	0.0213	1.6343	1.2784
No log	17.0	17	1.4628	0.0213	1.4628	1.2095
No log	18.0	18	1.3264	0.0213	1.3264	1.1517
No log	19.0	19	1.2150	0.0213	1.2150	1.1023
No log	20.0	20	1.1067	0.0213	1.1067	1.0520
No log	21.0	21	1.0178	0.0213	1.0178	1.0088
No log	22.0	22	0.9329	0.0213	0.9329	0.9658
No log	23.0	23	0.8707	0.1669	0.8707	0.9331
No log	24.0	24	0.8241	0.3771	0.8241	0.9078
No log	25.0	25	0.7743	0.3308	0.7743	0.8800
No log	26.0	26	0.7496	0.2912	0.7496	0.8658
No log	27.0	27	0.7113	0.3251	0.7113	0.8434
No log	28.0	28	0.6801	0.3540	0.6801	0.8247
No log	29.0	29	0.6643	0.3466	0.6643	0.8151
No log	30.0	30	0.6584	0.3438	0.6584	0.8114
No log	31.0	31	0.6199	0.3811	0.6199	0.7873
No log	32.0	32	0.6020	0.4105	0.6020	0.7759
No log	33.0	33	0.5844	0.4567	0.5844	0.7645
No log	34.0	34	0.5723	0.4706	0.5723	0.7565
No log	35.0	35	0.5670	0.4910	0.5670	0.7530
No log	36.0	36	0.5592	0.5117	0.5592	0.7478
No log	37.0	37	0.5568	0.5520	0.5568	0.7462
No log	38.0	38	0.5616	0.6050	0.5616	0.7494
No log	39.0	39	0.5621	0.6059	0.5621	0.7498
No log	40.0	40	0.5688	0.6337	0.5688	0.7542
No log	41.0	41	0.5874	0.6159	0.5874	0.7664
No log	42.0	42	0.5995	0.6326	0.5995	0.7743
No log	43.0	43	0.5999	0.6455	0.5999	0.7745
No log	44.0	44	0.6353	0.6201	0.6353	0.7971
No log	45.0	45	0.6205	0.6525	0.6205	0.7877
No log	46.0	46	0.6312	0.6394	0.6312	0.7945
No log	47.0	47	0.6711	0.6308	0.6711	0.8192
No log	48.0	48	0.6903	0.6191	0.6903	0.8309
No log	49.0	49	0.6519	0.6436	0.6519	0.8074
No log	50.0	50	0.6756	0.6215	0.6756	0.8220
No log	51.0	51	0.9182	0.5861	0.9182	0.9582
No log	52.0	52	0.9408	0.5861	0.9408	0.9700
No log	53.0	53	0.7383	0.6174	0.7383	0.8592
No log	54.0	54	0.6588	0.6397	0.6588	0.8117
No log	55.0	55	0.6544	0.6428	0.6544	0.8089
No log	56.0	56	0.7192	0.6270	0.7192	0.8480
No log	57.0	57	1.0600	0.5473	1.0600	1.0296
No log	58.0	58	1.1876	0.5203	1.1876	1.0898
No log	59.0	59	1.0262	0.5594	1.0262	1.0130
No log	60.0	60	0.7374	0.6178	0.7374	0.8587
No log	61.0	61	0.6633	0.6467	0.6633	0.8144
No log	62.0	62	0.6889	0.6387	0.6889	0.8300
No log	63.0	63	0.7196	0.6413	0.7196	0.8483
No log	64.0	64	0.8457	0.6052	0.8457	0.9196
No log	65.0	65	0.9160	0.5845	0.9159	0.9571

Framework versions

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