Version_weird_ASAP_FineTuningBERT_AugV12_k2_task1_organization_k2_k2_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: 0.8212
• Qwk: 0.6020
• Mse: 0.8212
• Rmse: 0.9062

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	12.0953	-0.0069	12.0953	3.4778
No log	2.0	2	10.7527	-0.0059	10.7527	3.2791
No log	3.0	3	9.4586	0.0018	9.4586	3.0755
No log	4.0	4	8.4742	0.0018	8.4742	2.9110
No log	5.0	5	7.4879	0.0018	7.4879	2.7364
No log	6.0	6	6.6788	0.0018	6.6788	2.5843
No log	7.0	7	6.1203	0.0011	6.1203	2.4739
No log	8.0	8	5.3684	0.0218	5.3684	2.3170
No log	9.0	9	4.7008	0.0156	4.7008	2.1681
No log	10.0	10	4.1880	0.0156	4.1880	2.0465
No log	11.0	11	3.7247	0.0079	3.7247	1.9299
No log	12.0	12	3.3374	0.0040	3.3374	1.8268
No log	13.0	13	2.9040	0.0040	2.9040	1.7041
No log	14.0	14	2.5569	0.0040	2.5569	1.5990
No log	15.0	15	2.2934	0.1594	2.2934	1.5144
No log	16.0	16	2.0637	0.0828	2.0637	1.4366
No log	17.0	17	1.8564	0.0470	1.8564	1.3625
No log	18.0	18	1.6732	0.0420	1.6732	1.2935
No log	19.0	19	1.5034	0.0420	1.5034	1.2261
No log	20.0	20	1.3516	0.0316	1.3516	1.1626
No log	21.0	21	1.2316	0.0212	1.2316	1.1098
No log	22.0	22	1.1272	0.0212	1.1272	1.0617
No log	23.0	23	1.0321	0.0316	1.0321	1.0159
No log	24.0	24	0.9525	0.0316	0.9525	0.9760
No log	25.0	25	0.8839	0.0316	0.8839	0.9401
No log	26.0	26	0.8420	0.2901	0.8420	0.9176
No log	27.0	27	0.7914	0.4234	0.7914	0.8896
No log	28.0	28	0.7405	0.4138	0.7405	0.8605
No log	29.0	29	0.6910	0.4382	0.6910	0.8312
No log	30.0	30	0.6506	0.4538	0.6506	0.8066
No log	31.0	31	0.6117	0.4850	0.6117	0.7821
No log	32.0	32	0.6034	0.4367	0.6034	0.7768
No log	33.0	33	0.5483	0.4977	0.5483	0.7405
No log	34.0	34	0.5232	0.5550	0.5232	0.7233
No log	35.0	35	0.5087	0.5236	0.5087	0.7132
No log	36.0	36	0.5490	0.4795	0.5490	0.7410
No log	37.0	37	0.5293	0.4985	0.5293	0.7275
No log	38.0	38	0.4912	0.5391	0.4912	0.7009
No log	39.0	39	0.5025	0.5301	0.5025	0.7089
No log	40.0	40	0.5651	0.5069	0.5651	0.7517
No log	41.0	41	0.5411	0.5875	0.5411	0.7356
No log	42.0	42	0.5293	0.6205	0.5293	0.7275
No log	43.0	43	0.5578	0.6209	0.5578	0.7469
No log	44.0	44	0.6621	0.5638	0.6621	0.8137
No log	45.0	45	0.6520	0.5776	0.6520	0.8075
No log	46.0	46	0.5829	0.6356	0.5829	0.7635
No log	47.0	47	0.6412	0.6035	0.6412	0.8008
No log	48.0	48	0.7585	0.5489	0.7585	0.8709
No log	49.0	49	0.7499	0.5579	0.7499	0.8659
No log	50.0	50	0.6637	0.6129	0.6637	0.8147
No log	51.0	51	0.6978	0.6026	0.6978	0.8353
No log	52.0	52	0.8705	0.5340	0.8705	0.9330
No log	53.0	53	0.8617	0.5487	0.8617	0.9283
No log	54.0	54	0.7155	0.6181	0.7155	0.8459
No log	55.0	55	0.7258	0.6154	0.7258	0.8520
No log	56.0	56	0.8755	0.5541	0.8755	0.9357
No log	57.0	57	0.9762	0.5308	0.9762	0.9880
No log	58.0	58	0.9445	0.5466	0.9445	0.9719
No log	59.0	59	0.8313	0.5914	0.8313	0.9118
No log	60.0	60	0.9198	0.5642	0.9198	0.9590
No log	61.0	61	1.0846	0.5299	1.0846	1.0414
No log	62.0	62	1.0749	0.5347	1.0749	1.0368
No log	63.0	63	1.0902	0.5362	1.0902	1.0441
No log	64.0	64	1.2252	0.5054	1.2252	1.1069
No log	65.0	65	1.1047	0.5252	1.1047	1.0510
No log	66.0	66	0.8212	0.6020	0.8212	0.9062

Framework versions

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