mix_ensemble_super_long_v1

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2879
• Accuracy Offensive: 0.9184
• F1 Offensive: 0.9157
• Accuracy Targeted: 0.9226
• F1 Targeted: 0.9006
• Accuracy Stance: 0.8693
• F1 Stance: 0.8271

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.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: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Offensive	F1 Offensive	Accuracy Targeted	F1 Targeted	Accuracy Stance	F1 Stance
0.7637	1.0	1324	0.7555	0.6545	0.5178	0.6545	0.5178	0.7009	0.5777
0.7299	2.0	2648	0.7125	0.6639	0.5465	0.6556	0.5204	0.7009	0.5777
0.7024	3.0	3972	0.6648	0.6998	0.6314	0.7100	0.6593	0.7017	0.5813
0.6657	4.0	5296	0.6314	0.7107	0.6451	0.7368	0.6994	0.7273	0.6532
0.6458	5.0	6620	0.5971	0.7508	0.7129	0.7704	0.7474	0.7515	0.7056
0.6336	6.0	7944	0.5819	0.7300	0.6731	0.7749	0.7434	0.7606	0.7057
0.6016	7.0	9268	0.5498	0.7674	0.7337	0.7980	0.7763	0.7738	0.7302
0.5853	8.0	10592	0.5281	0.7742	0.7421	0.8172	0.7955	0.7829	0.7404
0.5675	9.0	11916	0.5150	0.7545	0.7084	0.8229	0.7978	0.7968	0.7485
0.5497	10.0	13240	0.4831	0.8104	0.7894	0.8501	0.8304	0.8063	0.7673
0.5315	11.0	14564	0.4642	0.7987	0.7730	0.8550	0.8330	0.8127	0.7684
0.5342	12.0	15888	0.4416	0.8089	0.7864	0.8693	0.8480	0.8270	0.7840
0.5177	13.0	17212	0.4280	0.8350	0.8200	0.8784	0.8576	0.8319	0.7903
0.5035	14.0	18536	0.4040	0.8433	0.8301	0.8920	0.8709	0.8353	0.7950
0.4983	15.0	19860	0.3904	0.8433	0.8296	0.8999	0.8785	0.8489	0.8059
0.4837	16.0	21184	0.3985	0.8063	0.7815	0.8890	0.8666	0.8391	0.7926
0.4844	17.0	22508	0.3625	0.8667	0.8574	0.9082	0.8866	0.8554	0.8127
0.4691	18.0	23832	0.3616	0.8633	0.8533	0.9060	0.8841	0.8520	0.8082
0.4541	19.0	25156	0.3479	0.8882	0.8824	0.9116	0.8900	0.8573	0.8156
0.45	20.0	26480	0.3413	0.8682	0.8590	0.9139	0.8919	0.8633	0.8195
0.4427	21.0	27804	0.3356	0.8939	0.8889	0.9162	0.8945	0.8569	0.8159
0.4281	22.0	29128	0.3259	0.8705	0.8615	0.9184	0.8963	0.8603	0.8156
0.4408	23.0	30452	0.3162	0.8901	0.8842	0.9222	0.9001	0.8663	0.8223
0.4469	24.0	31776	0.3143	0.9128	0.9095	0.9215	0.8997	0.8633	0.8221
0.4115	25.0	33100	0.3104	0.9196	0.9170	0.9177	0.8960	0.8614	0.8208
0.4231	26.0	34424	0.3026	0.9154	0.9125	0.9237	0.9017	0.8614	0.8199
0.4224	27.0	35748	0.2949	0.9094	0.9057	0.9290	0.9068	0.8682	0.8245
0.4169	28.0	37072	0.2830	0.9248	0.9227	0.9286	0.9065	0.8708	0.8292
0.4128	29.0	38396	0.2935	0.9222	0.9198	0.9230	0.9010	0.8667	0.8243
0.4103	30.0	39720	0.2870	0.9267	0.9248	0.9226	0.9007	0.8629	0.8220

Framework versions

• Transformers 4.50.2
• Pytorch 2.6.0+cu124
• Datasets 3.0.1
• Tokenizers 0.21.1