Version3_ASAP_FineTuningBERT_AugV12_k1_task1_organization_k1_k1_fold2

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: 0.6795
• Qwk: 0.5460
• Mse: 0.6793
• Rmse: 0.8242

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	9.2162	0.0	9.2166	3.0359
No log	2.0	4	8.2870	0.0	8.2873	2.8788
No log	3.0	6	7.0289	0.0	7.0291	2.6513
No log	4.0	8	5.7733	0.0623	5.7736	2.4028
No log	5.0	10	4.7573	0.0	4.7576	2.1812
No log	6.0	12	4.0067	0.0	4.0071	2.0018
No log	7.0	14	3.2408	0.0	3.2412	1.8003
No log	8.0	16	2.5657	0.0	2.5661	1.6019
No log	9.0	18	2.0549	0.0665	2.0553	1.4336
No log	10.0	20	1.6343	0.0107	1.6348	1.2786
No log	11.0	22	1.3372	0.0	1.3377	1.1566
No log	12.0	24	1.1203	0.0	1.1207	1.0586
No log	13.0	26	0.9478	0.0	0.9482	0.9738
No log	14.0	28	0.8230	0.4023	0.8233	0.9074
No log	15.0	30	0.7254	0.4232	0.7257	0.8519
No log	16.0	32	0.6006	0.4798	0.6007	0.7751
No log	17.0	34	0.5426	0.5351	0.5427	0.7367
No log	18.0	36	0.5185	0.5603	0.5186	0.7202
No log	19.0	38	0.6575	0.5292	0.6577	0.8110
No log	20.0	40	0.4827	0.5476	0.4827	0.6948
No log	21.0	42	0.4380	0.5656	0.4380	0.6618
No log	22.0	44	0.6268	0.5250	0.6270	0.7918
No log	23.0	46	0.4378	0.5540	0.4379	0.6618
No log	24.0	48	0.5257	0.5116	0.5258	0.7251
No log	25.0	50	0.5988	0.4482	0.5989	0.7739
No log	26.0	52	0.6026	0.4725	0.6026	0.7763
No log	27.0	54	0.5033	0.5905	0.5034	0.7095
No log	28.0	56	0.7774	0.4707	0.7773	0.8817
No log	29.0	58	0.5842	0.5668	0.5842	0.7643
No log	30.0	60	0.7053	0.4990	0.7051	0.8397
No log	31.0	62	0.7444	0.4921	0.7440	0.8626
No log	32.0	64	0.7124	0.4980	0.7120	0.8438
No log	33.0	66	0.6906	0.4891	0.6902	0.8308
No log	34.0	68	0.7415	0.4708	0.7411	0.8609
No log	35.0	70	0.7536	0.4140	0.7531	0.8678
No log	36.0	72	0.7869	0.3921	0.7865	0.8869
No log	37.0	74	0.8030	0.3896	0.8027	0.8959
No log	38.0	76	0.7640	0.4287	0.7638	0.8739
No log	39.0	78	0.7232	0.4563	0.7230	0.8503
No log	40.0	80	0.7644	0.4063	0.7643	0.8742
No log	41.0	82	0.8053	0.3835	0.8052	0.8973
No log	42.0	84	0.8455	0.4149	0.8454	0.9194
No log	43.0	86	0.7926	0.4417	0.7925	0.8902
No log	44.0	88	0.8038	0.4878	0.8036	0.8964
No log	45.0	90	0.7471	0.5357	0.7469	0.8642
No log	46.0	92	0.6811	0.5262	0.6810	0.8252
No log	47.0	94	0.6795	0.5460	0.6793	0.8242

Framework versions

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