learning_rate: 5e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.06
num_epochs: 15
label_smoothing_factor: 0.3

Training results

Training Loss	Epoch	Step	Validation Loss	Sari	Sari Add	Sari Keep	Sari Del	Fkgl	Bleu	D Sari	D Sari Keep	D Sari Del	D Sari Add
6.1328	0.0458	500	5.3138	43.1674	0.2987	42.131	87.0726	8.6097	14.0105	0.3408	0.2745	0.7195	0.0283
5.3835	0.0915	1000	5.1584	44.2915	1.6844	43.9062	87.2839	8.4261	15.8124	0.3674	0.305	0.7466	0.0504
5.2631	0.1373	1500	5.0900	43.998	3.132	43.426	85.4358	8.3225	20.3359	0.3749	0.3161	0.734	0.0746
5.2175	0.1830	2000	5.0468	44.9432	4.254	44.2329	86.3427	8.0816	19.5926	0.3866	0.3266	0.7426	0.0906
5.1732	0.2288	2500	5.0134	45.7478	4.8691	44.5363	87.8379	7.6111	16.4266	0.4008	0.3383	0.7689	0.0952
5.1462	0.2745	3000	4.9859	45.7366	4.895	45.0612	87.2536	7.6222	18.5103	0.399	0.3399	0.7579	0.0992
5.101	0.3203	3500	4.9758	46.4824	6.0787	44.7909	88.5775	7.1986	15.1778	0.4105	0.3459	0.7774	0.1082
5.0871	0.3660	4000	4.9549	46.0324	5.4413	44.7164	87.9396	7.4182	16.5486	0.4061	0.342	0.77	0.1063
5.0729	0.4118	4500	4.9537	46.293	5.77	44.2683	88.8407	7.1858	13.4431	0.4147	0.3465	0.7875	0.1099
5.0496	0.4575	5000	4.9436	46.4777	7.0556	45.253	87.1245	7.097	20.3347	0.4132	0.3523	0.7613	0.1259
5.0582	0.5033	5500	4.9383	47.3154	7.242	46.2796	88.4246	7.0127	17.5809	0.426	0.3654	0.779	0.1335
5.0497	0.5490	6000	4.9252	47.2003	6.6904	45.8378	89.0727	7.0475	14.6809	0.4237	0.3575	0.7879	0.1257
5.025	0.5948	6500	4.9213	46.0976	7.1373	45.0427	86.1127	7.7905	22.4406	0.4136	0.3527	0.755	0.1332
5.012	0.6406	7000	4.9229	47.234	7.4491	45.7379	88.515	7.011	16.9018	0.4264	0.3593	0.7858	0.134
5.0143	0.6863	7500	4.9119	46.8084	7.464	44.8179	88.1433	7.3154	17.2521	0.422	0.3548	0.7805	0.1307
5.0059	0.7321	8000	4.9099	45.5319	6.6394	44.4112	85.5451	7.896	22.6684	0.4033	0.3455	0.7349	0.1294
5.0207	0.7778	8500	4.9063	47.5768	8.1527	45.5744	89.0034	6.7988	15.693	0.4315	0.3656	0.7916	0.1373
5.0091	0.8236	9000	4.9095	47.0639	7.4532	44.7893	88.9493	7.0553	14.5321	0.4274	0.3575	0.7917	0.1329
5.006	0.8693	9500	4.9060	46.3942	7.1108	45.8815	86.1904	7.3552	23.1795	0.4185	0.3594	0.7611	0.1351
4.9942	0.9151	10000	4.8999	47.0098	7.171	45.4075	88.4509	6.9761	16.5593	0.4285	0.3685	0.7853	0.1316
4.9758	0.9608	10500	4.9013	47.0887	7.9423	46.0327	87.2912	7.3689	21.0823	0.4257	0.3656	0.7719	0.1397
4.9542	1.0066	11000	4.8895	47.3872	8.4177	45.7822	87.9617	7.2673	19.2194	0.4312	0.368	0.781	0.1445
4.9219	1.0523	11500	4.8790	46.3681	7.7671	44.7413	86.5959	7.6345	21.2783	0.4252	0.3629	0.7706	0.1421
4.9112	1.0981	12000	4.8838	47.027	8.1625	45.5826	87.3361	7.258	20.6417	0.4298	0.3691	0.7747	0.1456
4.911	1.1439	12500	4.8831	47.1226	7.8491	45.4658	88.0527	7.2421	18.1067	0.4303	0.3652	0.7838	0.1419
4.9152	1.1896	13000	4.8855	47.1375	7.8122	45.3166	88.2838	7.2476	17.2915	0.4302	0.3646	0.7882	0.1378
4.9095	1.2354	13500	4.8688	47.2743	8.0396	45.4407	88.3427	7.1222	17.3563	0.432	0.3675	0.7871	0.1412
4.9059	1.2811	14000	4.8668	47.6801	8.4487	46.1395	88.4521	6.86	18.0804	0.4344	0.3716	0.7864	0.145
4.9041	1.3269	14500	4.8702	47.1864	7.999	45.9918	87.5684	7.4014	20.17	0.4297	0.3676	0.7784	0.1431
4.8981	1.3726	15000	4.8723	47.2509	8.3204	46.1136	87.3186	7.4238	21.2725	0.4254	0.3648	0.7687	0.1427
4.9084	1.4184	15500	4.8603	46.8462	8.7682	45.5791	86.1913	7.3917	23.7804	0.4229	0.3606	0.7566	0.1514
4.8936	1.4641	16000	4.8629	46.7335	8.1781	45.4662	86.5561	7.6962	22.3547	0.4224	0.3623	0.7597	0.1453
4.8915	1.5099	16500	4.8574	47.4648	8.1403	46.24	88.014	7.2278	19.176	0.4254	0.3623	0.7701	0.1436
4.8825	1.5556	17000	4.8599	47.6965	8.4756	46.4106	88.2032	7.2406	19.1069	0.4329	0.3697	0.7812	0.1479
4.8889	1.6014	17500	4.8654	48.0375	8.5524	46.8612	88.6988	6.8238	18.0395	0.4385	0.3749	0.7925	0.1481
4.8839	1.6471	18000	4.8545	47.4889	8.6392	46.4123	87.4151	7.3073	21.379	0.4342	0.374	0.7724	0.1563
4.8741	1.6929	18500	4.8509	47.7652	8.2965	46.1359	88.863	6.9654	16.5738	0.4353	0.368	0.7908	0.1472
4.8995	1.7387	19000	4.8438	47.6134	8.0189	46.6421	88.1792	7.1982	18.9587	0.4338	0.3689	0.7844	0.148
4.8731	1.7844	19500	4.8409	47.4026	8.2937	45.8033	88.1108	7.1948	18.7212	0.4313	0.3673	0.7803	0.1463
4.89	1.8302	20000	4.8400	47.5569	8.5899	46.2326	87.8483	7.291	19.9641	0.4348	0.3721	0.7792	0.1531
4.8933	1.8759	20500	4.8421	47.7037	8.9349	46.6601	87.5161	7.2229	21.6357	0.4305	0.3687	0.7706	0.1522
4.8721	1.9217	21000	4.8352	47.6211	8.3506	46.0772	88.4355	7.2742	17.8295	0.4373	0.3709	0.7894	0.1516
4.8843	1.9674	21500	4.8295	47.7593	8.5392	45.897	88.8416	7.1369	16.4039	0.4326	0.3653	0.7845	0.148
4.8644	2.0132	22000	4.8423	47.7889	9.2687	46.5503	87.5476	7.0183	21.7765	0.4367	0.3769	0.7742	0.1589
4.7805	2.0589	22500	4.8378	48.0466	9.1884	46.876	88.0753	6.9269	20.5624	0.433	0.3725	0.7743	0.1522

Framework versions

Transformers 4.57.3
Pytorch 2.9.1+cu128
Datasets 3.6.0
Tokenizers 0.22.1

taiypeo/bart-large-d-wikipedia

bart-large-d-wikipedia

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

taiypeo/
bart-large-d-wikipedia