Version_weird_ASAP_FineTuningBERT_AugV12_k4_task1_organization_k4_k4_fold0

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: 0.4921
• Qwk: 0.6538
• Mse: 0.4921
• Rmse: 0.7015

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	9.0472	0.0	9.0472	3.0079
No log	2.0	4	7.9221	0.0	7.9221	2.8146
No log	3.0	6	6.8886	0.0	6.8886	2.6246
No log	4.0	8	6.3752	-0.0104	6.3752	2.5249
No log	5.0	10	5.3922	0.0115	5.3922	2.3221
No log	6.0	12	4.6700	0.0039	4.6700	2.1610
No log	7.0	14	3.8905	0.0	3.8905	1.9724
No log	8.0	16	3.2734	0.0	3.2734	1.8092
No log	9.0	18	2.5413	0.0249	2.5413	1.5942
No log	10.0	20	2.0803	0.0484	2.0803	1.4423
No log	11.0	22	1.5568	0.0316	1.5568	1.2477
No log	12.0	24	1.2682	0.0316	1.2682	1.1261
No log	13.0	26	1.0577	0.0316	1.0577	1.0285
No log	14.0	28	0.8903	0.0638	0.8903	0.9436
No log	15.0	30	0.7871	0.3488	0.7871	0.8872
No log	16.0	32	0.7494	0.1397	0.7494	0.8657
No log	17.0	34	0.7655	0.1291	0.7655	0.8749
No log	18.0	36	0.6677	0.2309	0.6677	0.8171
No log	19.0	38	0.6409	0.2436	0.6409	0.8006
No log	20.0	40	0.9014	0.1502	0.9014	0.9494
No log	21.0	42	0.9363	0.3003	0.9363	0.9676
No log	22.0	44	0.6304	0.4303	0.6304	0.7940
No log	23.0	46	0.7385	0.4585	0.7385	0.8593
No log	24.0	48	0.8764	0.4372	0.8764	0.9362
No log	25.0	50	0.5612	0.5366	0.5612	0.7491
No log	26.0	52	0.4955	0.6256	0.4955	0.7039
No log	27.0	54	0.6483	0.6093	0.6483	0.8052
No log	28.0	56	0.6415	0.6329	0.6415	0.8009
No log	29.0	58	0.5506	0.6534	0.5506	0.7420
No log	30.0	60	0.6163	0.6178	0.6163	0.7851
No log	31.0	62	0.4760	0.6746	0.4760	0.6899
No log	32.0	64	0.5253	0.6486	0.5253	0.7248
No log	33.0	66	0.6693	0.6318	0.6693	0.8181
No log	34.0	68	0.6420	0.6095	0.6420	0.8013
No log	35.0	70	0.6882	0.5755	0.6882	0.8296
No log	36.0	72	0.6548	0.6233	0.6548	0.8092
No log	37.0	74	0.7262	0.5965	0.7262	0.8522
No log	38.0	76	0.4442	0.6361	0.4442	0.6665
No log	39.0	78	0.4526	0.5938	0.4526	0.6728
No log	40.0	80	0.5555	0.6121	0.5555	0.7453
No log	41.0	82	0.6148	0.6028	0.6148	0.7841
No log	42.0	84	0.4387	0.6176	0.4387	0.6624
No log	43.0	86	0.4555	0.6156	0.4555	0.6749
No log	44.0	88	0.5768	0.6276	0.5768	0.7594
No log	45.0	90	0.6569	0.6274	0.6569	0.8105
No log	46.0	92	0.4778	0.6456	0.4778	0.6913
No log	47.0	94	0.4846	0.6254	0.4846	0.6962
No log	48.0	96	0.5414	0.6381	0.5414	0.7358
No log	49.0	98	0.6347	0.6042	0.6347	0.7967
No log	50.0	100	0.4861	0.6464	0.4861	0.6972
No log	51.0	102	0.4921	0.6538	0.4921	0.7015

Framework versions

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