t5-small-fine-tuned_model_3

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1560
• Rouge1: 75.4228
• Rouge2: 70.7071
• Rougel: 74.0159
• Rougelsum: 74.2555
• Gen Len: 396.1667

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: 0.005
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	0.7692	10	1.7251	13.6123	5.7258	13.1801	13.1787	1027.0
No log	1.5385	20	1.5442	23.3131	17.9737	23.3131	23.3131	1027.0
No log	2.3077	30	1.3803	12.2977	5.1904	11.7431	11.5969	1027.0
No log	3.0769	40	1.2344	13.8993	11.771	13.8993	14.0091	1027.0
No log	3.8462	50	1.1516	13.9042	11.8938	13.9042	14.0103	1027.0
No log	4.6154	60	0.9481	14.3687	10.4481	13.1919	13.0214	876.6667
No log	5.3846	70	0.9286	27.0525	14.6747	24.9583	24.3756	857.5
No log	6.1538	80	0.8804	21.6353	12.8534	18.4791	18.4306	877.1667
No log	6.9231	90	0.7841	47.5579	30.341	42.5411	42.7482	550.5
No log	7.6923	100	0.7793	35.2203	25.009	31.3145	30.6642	859.8333
No log	8.4615	110	0.6860	37.2436	29.1438	33.1425	32.729	859.6667
No log	9.2308	120	0.7150	29.122	23.7579	27.5853	26.5771	859.6667
No log	10.0	130	0.6579	51.6814	37.1169	47.9067	47.9272	530.1667
No log	10.7692	140	0.6267	37.5717	28.0617	32.827	32.592	860.5
No log	11.5385	150	0.6118	62.1203	49.9121	55.4072	54.9256	564.0
No log	12.3077	160	0.5481	61.2435	49.738	55.7893	55.6371	565.1667
No log	13.0769	170	0.5685	57.4855	47.8398	54.6011	53.7537	407.8333
No log	13.8462	180	0.5603	63.7808	52.0648	58.9732	59.1514	107.6667
No log	14.6154	190	0.4906	56.541	43.5496	50.0309	49.4554	402.3333
No log	15.3846	200	0.4920	44.085	31.8595	41.9242	42.2744	130.6667
No log	16.1538	210	0.4519	57.8642	47.346	53.4872	53.7607	294.0
No log	16.9231	220	0.4319	44.5213	29.3385	36.2116	36.1914	481.0
No log	17.6923	230	0.4147	52.262	33.4537	42.1175	42.8641	335.6667
No log	18.4615	240	0.4411	33.5609	21.155	26.7958	27.0263	785.1667
No log	19.2308	250	0.3791	62.4765	48.3805	56.5917	55.6696	301.6667
No log	20.0	260	0.3913	66.6348	54.4823	59.6097	59.9255	144.5
No log	20.7692	270	0.3530	54.5169	46.9471	50.0583	49.4563	431.1667
No log	21.5385	280	0.3245	46.7808	38.4793	42.4197	42.2085	712.3333
No log	22.3077	290	0.3368	47.0382	35.8428	39.104	38.5503	735.8333
No log	23.0769	300	0.3297	53.7986	44.0834	46.4405	47.5762	654.3333
No log	23.8462	310	0.2940	59.8414	45.4853	53.3007	53.3967	155.8333
No log	24.6154	320	0.3340	65.9	52.8727	59.5371	59.484	227.1667
No log	25.3846	330	0.2812	58.7644	47.4464	51.8233	52.0057	302.5
No log	26.1538	340	0.2787	64.4588	51.5866	56.3922	55.9368	219.8333
No log	26.9231	350	0.2872	55.727	45.0152	49.2849	49.2548	601.0
No log	27.6923	360	0.2971	63.8289	52.3489	57.6671	57.2489	361.0
No log	28.4615	370	0.2893	60.4914	49.2527	54.2347	54.2306	174.1667
No log	29.2308	380	0.2479	65.7383	53.9204	57.7956	58.0014	304.0
No log	30.0	390	0.2452	58.2415	49.1706	49.5983	49.1554	630.1667
No log	30.7692	400	0.2504	54.9945	42.7543	45.5489	46.7113	664.3333
No log	31.5385	410	0.2361	62.8874	47.848	52.1486	52.6791	439.3333
No log	32.3077	420	0.2282	35.307	20.7981	25.3321	25.7283	648.3333
No log	33.0769	430	0.2268	39.9343	26.2938	32.5539	32.5389	464.8333
No log	33.8462	440	0.2160	37.5551	29.1716	36.4583	36.3205	23.0
No log	34.6154	450	0.2049	43.1026	33.2667	40.7167	40.7024	108.0
No log	35.3846	460	0.2006	61.876	50.0227	53.1594	53.2425	502.6667
No log	36.1538	470	0.1934	60.7038	50.0727	55.2509	54.8126	338.8333
No log	36.9231	480	0.1960	70.3567	56.2927	61.7649	62.2948	358.6667
No log	37.6923	490	0.1792	59.3192	42.9024	47.1844	47.5165	355.5
0.5531	38.4615	500	0.1755	58.8161	44.5037	47.7178	47.6386	501.5
0.5531	39.2308	510	0.1892	54.0773	43.7896	47.246	47.0727	440.1667
0.5531	40.0	520	0.1821	57.2344	46.5657	52.5641	52.5542	589.1667
0.5531	40.7692	530	0.1729	68.5089	53.586	60.131	60.3304	292.6667
0.5531	41.5385	540	0.1989	63.9246	51.624	55.4652	55.8813	355.3333
0.5531	42.3077	550	0.1868	60.7441	50.1997	55.0352	53.7644	564.3333
0.5531	43.0769	560	0.1570	44.0831	33.923	37.6398	37.451	748.6667
0.5531	43.8462	570	0.1806	60.5725	47.5269	52.2245	53.3507	487.8333
0.5531	44.6154	580	0.1984	64.7623	56.5668	58.7952	59.3482	527.1667
0.5531	45.3846	590	0.1673	62.8231	50.6443	53.4276	53.4813	385.8333
0.5531	46.1538	600	0.1593	77.1493	70.2538	73.9133	74.0634	336.5
0.5531	46.9231	610	0.1787	69.6579	57.144	62.8631	63.1825	264.1667
0.5531	47.6923	620	0.1579	67.3991	55.4929	60.496	59.9907	237.5
0.5531	48.4615	630	0.1510	55.7614	52.4735	54.2066	54.4553	351.3333
0.5531	49.2308	640	0.1490	66.8343	59.1175	62.6098	62.6185	489.1667
0.5531	50.0	650	0.1450	73.7447	68.8381	72.2138	71.7347	403.1667
0.5531	50.7692	660	0.1435	73.4612	62.1625	67.6424	67.8374	335.0
0.5531	51.5385	670	0.1412	69.9245	63.2467	67.5193	66.7139	459.3333
0.5531	52.3077	680	0.1537	67.309	56.0056	60.5465	60.7674	483.3333
0.5531	53.0769	690	0.1618	66.0585	54.5418	60.2616	59.8329	391.1667
0.5531	53.8462	700	0.1546	62.9813	57.9394	61.4801	60.8618	532.5
0.5531	54.6154	710	0.1768	69.2968	62.2167	65.5068	65.6779	463.5
0.5531	55.3846	720	0.1523	70.6019	64.4629	68.7182	68.6705	468.3333
0.5531	56.1538	730	0.1452	74.6336	70.8117	73.3083	73.5846	427.5
0.5531	56.9231	740	0.1458	80.2581	73.4241	77.8048	78.2945	321.5
0.5531	57.6923	750	0.1454	69.5709	60.7631	64.0057	64.1665	438.5
0.5531	58.4615	760	0.1440	74.8974	70.6795	73.4561	73.6899	415.6667
0.5531	59.2308	770	0.1420	75.8343	70.7545	74.2487	74.3303	370.8333
0.5531	60.0	780	0.1518	68.975	60.6509	63.3542	63.4528	488.0
0.5531	60.7692	790	0.1329	75.4609	65.9764	70.407	70.9722	379.6667
0.5531	61.5385	800	0.1298	75.6475	67.6634	72.3407	72.6996	405.3333
0.5531	62.3077	810	0.1324	76.1183	68.3992	73.0096	73.3558	379.3333
0.5531	63.0769	820	0.1469	61.1852	57.2433	60.7155	60.4608	675.1667
0.5531	63.8462	830	0.1385	68.2356	60.6576	63.8079	63.9332	513.1667
0.5531	64.6154	840	0.1434	71.3804	66.5798	69.5366	69.5204	508.0
0.5531	65.3846	850	0.1557	63.2252	59.4299	61.8559	61.89	537.0
0.5531	66.1538	860	0.1489	74.2213	68.7578	72.1378	72.1929	472.1667
0.5531	66.9231	870	0.1582	79.3572	72.5039	77.4724	77.8716	324.5
0.5531	67.6923	880	0.1419	70.4109	65.0778	68.5519	68.6548	523.0
0.5531	68.4615	890	0.1403	75.0692	67.5111	72.954	73.2228	379.3333
0.5531	69.2308	900	0.1411	74.8948	66.439	72.7139	73.0614	383.0
0.5531	70.0	910	0.1423	79.3572	71.8921	77.4724	77.8716	325.5
0.5531	70.7692	920	0.1398	79.3572	72.135	77.4724	77.8716	325.5
0.5531	71.5385	930	0.1376	75.2809	70.7071	73.6409	73.8805	410.0
0.5531	72.3077	940	0.1440	75.7518	70.6157	74.2567	74.4963	381.0
0.5531	73.0769	950	0.1434	80.9338	73.4733	78.7226	79.3074	319.1667
0.5531	73.8462	960	0.1403	80.33	73.1042	78.1987	78.7715	321.0
0.5531	74.6154	970	0.1393	75.7518	70.7071	74.2151	74.4547	377.6667
0.5531	75.3846	980	0.1363	75.2169	70.6795	73.6694	73.9091	414.1667
0.5531	76.1538	990	0.1392	75.7518	70.7639	74.5831	74.8227	371.6667
0.0743	76.9231	1000	0.1457	75.8091	71.008	74.7065	74.9461	369.5
0.0743	77.6923	1010	0.1476	75.6793	70.7662	74.2724	74.512	389.0
0.0743	78.4615	1020	0.1504	74.9721	70.6949	73.5623	73.6876	419.8333
0.0743	79.2308	1030	0.1488	74.9721	70.6949	73.5623	73.6876	419.8333
0.0743	80.0	1040	0.1457	67.2012	63.9833	66.413	66.8448	518.6667
0.0743	80.7692	1050	0.1411	75.0783	70.1206	73.56	73.7876	416.8333
0.0743	81.5385	1060	0.1444	74.9181	70.6595	73.5353	73.7381	430.0
0.0743	82.3077	1070	0.1661	75.252	70.7071	73.6151	73.8548	412.0
0.0743	83.0769	1080	0.1686	75.7518	71.0652	74.5395	74.7791	367.6667
0.0743	83.8462	1090	0.1691	75.0598	70.7071	73.5513	73.7701	417.1667
0.0743	84.6154	1100	0.1678	74.9666	70.6637	73.4386	73.6548	423.0
0.0743	85.3846	1110	0.1671	74.7224	70.4484	73.3686	73.5149	453.0
0.0743	86.1538	1120	0.1656	75.7518	70.7717	74.3526	74.5922	378.3333
0.0743	86.9231	1130	0.1643	75.7518	70.7717	74.3526	74.5922	378.5
0.0743	87.6923	1140	0.1596	75.7518	70.7717	74.3526	74.5922	378.5
0.0743	88.4615	1150	0.1592	75.2818	70.7071	73.7514	73.991	403.1667
0.0743	89.2308	1160	0.1607	75.2883	70.7071	73.6474	73.887	410.3333
0.0743	90.0	1170	0.1600	75.0598	70.7071	73.5513	73.7701	417.1667
0.0743	90.7692	1180	0.1571	75.3879	70.7071	73.981	74.2206	397.0
0.0743	91.5385	1190	0.1561	75.3966	70.7071	73.9896	74.2292	396.8333
0.0743	92.3077	1200	0.1556	75.3794	70.7071	73.9724	74.2121	398.3333
0.0743	93.0769	1210	0.1555	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	93.8462	1220	0.1556	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	94.6154	1230	0.1557	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	95.3846	1240	0.1558	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	96.1538	1250	0.1559	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	96.9231	1260	0.1559	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	97.6923	1270	0.1559	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	98.4615	1280	0.1560	75.4228	70.7071	74.0159	74.2555	396.1667
0.0743	99.2308	1290	0.1560	75.7518	70.9213	74.5831	74.8227	370.3333
0.0743	100.0	1300	0.1560	75.4228	70.7071	74.0159	74.2555	396.1667

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1