Version_weird_ASAP_FineTuningBERT_AugV12_k1_task1_organization_k1_k1_fold3

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: 0.6616
• Qwk: 0.5976
• Mse: 0.6616
• Rmse: 0.8134

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	12.5206	0.0043	12.5194	3.5383
No log	2.0	2	10.1092	0.0065	10.1073	3.1792
No log	3.0	3	8.7336	0.0018	8.7318	2.9550
No log	4.0	4	7.2796	0.0018	7.2780	2.6978
No log	5.0	5	6.6110	0.0	6.6098	2.5710
No log	6.0	6	6.2344	-0.0062	6.2333	2.4967
No log	7.0	7	5.4486	0.0146	5.4475	2.3340
No log	8.0	8	4.6512	0.0038	4.6501	2.1564
No log	9.0	9	4.2589	0.0	4.2577	2.0634
No log	10.0	10	3.9553	0.0	3.9541	1.9885
No log	11.0	11	3.6181	0.0	3.6170	1.9018
No log	12.0	12	3.3234	0.0	3.3224	1.8227
No log	13.0	13	3.1030	0.0	3.1021	1.7613
No log	14.0	14	2.7762	0.0	2.7753	1.6659
No log	15.0	15	2.4382	0.1494	2.4373	1.5612
No log	16.0	16	2.1974	0.0452	2.1965	1.4821
No log	17.0	17	1.9284	0.0166	1.9275	1.3884
No log	18.0	18	1.6714	0.0102	1.6707	1.2925
No log	19.0	19	1.4427	0.0202	1.4421	1.2009
No log	20.0	20	1.2938	0.0202	1.2933	1.1372
No log	21.0	21	1.1687	0.0102	1.1682	1.0808
No log	22.0	22	1.1656	0.0102	1.1650	1.0794
No log	23.0	23	1.0645	0.0102	1.0640	1.0315
No log	24.0	24	0.9171	0.1636	0.9167	0.9575
No log	25.0	25	0.9547	0.2235	0.9544	0.9770
No log	26.0	26	0.8714	0.2544	0.8711	0.9333
No log	27.0	27	0.7828	0.3299	0.7825	0.8846
No log	28.0	28	0.7778	0.2534	0.7775	0.8817
No log	29.0	29	0.7499	0.2534	0.7496	0.8658
No log	30.0	30	0.6889	0.3335	0.6888	0.8299
No log	31.0	31	0.6533	0.3763	0.6534	0.8083
No log	32.0	32	0.6632	0.3334	0.6632	0.8144
No log	33.0	33	0.6900	0.3227	0.6901	0.8307
No log	34.0	34	0.6427	0.3627	0.6429	0.8018
No log	35.0	35	0.5639	0.4564	0.5641	0.7511
No log	36.0	36	0.5618	0.4928	0.5620	0.7497
No log	37.0	37	0.5416	0.4697	0.5418	0.7361
No log	38.0	38	0.6341	0.4388	0.6344	0.7965
No log	39.0	39	0.6703	0.4520	0.6706	0.8189
No log	40.0	40	0.6193	0.5587	0.6195	0.7871
No log	41.0	41	0.5207	0.5990	0.5209	0.7217
No log	42.0	42	0.5233	0.6131	0.5235	0.7235
No log	43.0	43	0.5282	0.6026	0.5285	0.7269
No log	44.0	44	0.5749	0.5796	0.5750	0.7583
No log	45.0	45	0.5573	0.5790	0.5575	0.7467
No log	46.0	46	0.5316	0.6287	0.5318	0.7292
No log	47.0	47	0.5377	0.6373	0.5379	0.7334
No log	48.0	48	0.5682	0.6257	0.5684	0.7539
No log	49.0	49	0.6202	0.5975	0.6204	0.7876
No log	50.0	50	0.5972	0.5998	0.5974	0.7729
No log	51.0	51	0.5694	0.6174	0.5695	0.7547
No log	52.0	52	0.5791	0.6165	0.5793	0.7611
No log	53.0	53	0.6258	0.5970	0.6259	0.7911
No log	54.0	54	0.5934	0.6079	0.5935	0.7704
No log	55.0	55	0.5530	0.6210	0.5531	0.7437
No log	56.0	56	0.5592	0.6073	0.5594	0.7479
No log	57.0	57	0.6072	0.6013	0.6073	0.7793
No log	58.0	58	0.6745	0.5924	0.6745	0.8213
No log	59.0	59	0.6736	0.5995	0.6737	0.8208
No log	60.0	60	0.6890	0.5982	0.6890	0.8301
No log	61.0	61	0.6258	0.6130	0.6258	0.7911
No log	62.0	62	0.5782	0.6227	0.5783	0.7605
No log	63.0	63	0.5510	0.6025	0.5511	0.7424
No log	64.0	64	0.5819	0.6174	0.5820	0.7629
No log	65.0	65	0.7051	0.5794	0.7050	0.8397
No log	66.0	66	0.7408	0.5637	0.7407	0.8606
No log	67.0	67	0.6616	0.5976	0.6616	0.8134

Framework versions

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