Version_weird_ASAP_FineTuningBERT_AugV12_k3_task1_organization_k3_k3_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: 0.6611
• Qwk: 0.6138
• Mse: 0.6604
• Rmse: 0.8126

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	11.7479	0.0249	11.7465	3.4273
No log	2.0	2	9.9811	0.0152	9.9800	3.1591
No log	3.0	3	8.4914	0.0018	8.4903	2.9138
No log	4.0	4	7.2570	0.0	7.2560	2.6937
No log	5.0	5	6.2626	0.0318	6.2618	2.5024
No log	6.0	6	5.6087	0.0331	5.6077	2.3681
No log	7.0	7	4.9755	0.0182	4.9748	2.2304
No log	8.0	8	4.3589	0.0059	4.3583	2.0876
No log	9.0	9	3.6190	0.0038	3.6183	1.9022
No log	10.0	10	3.0759	0.0	3.0751	1.7536
No log	11.0	11	2.6719	0.0096	2.6711	1.6343
No log	12.0	12	2.3323	0.1654	2.3316	1.5269
No log	13.0	13	2.1456	0.0890	2.1450	1.4646
No log	14.0	14	1.8946	0.0601	1.8940	1.3762
No log	15.0	15	1.6317	0.0488	1.6312	1.2772
No log	16.0	16	1.4064	0.0365	1.4059	1.1857
No log	17.0	17	1.2386	0.0365	1.2381	1.1127
No log	18.0	18	1.1194	0.0302	1.1189	1.0578
No log	19.0	19	1.0341	0.0302	1.0337	1.0167
No log	20.0	20	0.9471	0.0428	0.9467	0.9730
No log	21.0	21	0.8839	0.3368	0.8835	0.9399
No log	22.0	22	0.8108	0.3636	0.8104	0.9002
No log	23.0	23	0.7664	0.3136	0.7660	0.8752
No log	24.0	24	0.7416	0.2950	0.7412	0.8610
No log	25.0	25	0.7317	0.2768	0.7314	0.8552
No log	26.0	26	0.7200	0.2656	0.7197	0.8484
No log	27.0	27	0.7215	0.2642	0.7213	0.8493
No log	28.0	28	0.6982	0.2897	0.6980	0.8355
No log	29.0	29	0.6491	0.3114	0.6489	0.8055
No log	30.0	30	0.6518	0.3239	0.6516	0.8072
No log	31.0	31	0.6719	0.3276	0.6718	0.8196
No log	32.0	32	0.6649	0.3470	0.6648	0.8153
No log	33.0	33	0.5717	0.4422	0.5716	0.7560
No log	34.0	34	0.5669	0.5133	0.5668	0.7529
No log	35.0	35	0.5691	0.5806	0.5689	0.7542
No log	36.0	36	0.6201	0.5746	0.6199	0.7873
No log	37.0	37	0.5690	0.6135	0.5687	0.7541
No log	38.0	38	0.5023	0.6410	0.5020	0.7085
No log	39.0	39	0.5129	0.6505	0.5125	0.7159
No log	40.0	40	0.5603	0.6339	0.5599	0.7483
No log	41.0	41	0.5966	0.6279	0.5962	0.7722
No log	42.0	42	0.5802	0.6434	0.5798	0.7614
No log	43.0	43	0.5758	0.6447	0.5754	0.7585
No log	44.0	44	0.6264	0.6325	0.6259	0.7912
No log	45.0	45	0.7352	0.5951	0.7347	0.8571
No log	46.0	46	0.6963	0.6158	0.6957	0.8341
No log	47.0	47	0.6111	0.6287	0.6106	0.7814
No log	48.0	48	0.6169	0.6265	0.6163	0.7851
No log	49.0	49	0.6527	0.6104	0.6521	0.8075
No log	50.0	50	0.7616	0.6190	0.7610	0.8723
No log	51.0	51	0.7831	0.6156	0.7825	0.8846
No log	52.0	52	0.7063	0.6025	0.7056	0.8400
No log	53.0	53	0.6635	0.6121	0.6628	0.8141
No log	54.0	54	0.6600	0.6196	0.6592	0.8119
No log	55.0	55	0.7028	0.6208	0.7020	0.8379
No log	56.0	56	0.7564	0.6288	0.7556	0.8692
No log	57.0	57	0.8193	0.6104	0.8185	0.9047
No log	58.0	58	0.7167	0.6204	0.7159	0.8461
No log	59.0	59	0.6611	0.6138	0.6604	0.8126

Framework versions

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