t5-small-finetuned-xsum

This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2889
• Rouge1: 39.8328
• Rouge2: 22.4239
• Rougel: 39.9834
• Rougelsum: 39.9724
• Gen Len: 16.0805

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
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	70	0.2976	38.8006	21.9111	38.9213	38.9522	16.0847
No log	2.0	140	0.2965	38.5928	21.9111	38.7429	38.6634	16.1017
No log	3.0	210	0.2986	39.1757	22.0568	39.2841	39.3331	16.0847
No log	4.0	280	0.2962	39.4362	21.9778	39.4592	39.5481	16.0847
No log	5.0	350	0.2991	39.4154	22.3903	39.5864	39.574	16.0932
No log	6.0	420	0.2973	39.5908	22.5913	39.7026	39.7583	16.0932
No log	7.0	490	0.2973	39.6087	22.5315	39.7469	39.7952	16.0847
0.2763	8.0	560	0.2976	39.5607	22.5913	39.6733	39.7155	16.1102
0.2763	9.0	630	0.2986	38.7333	22.0903	38.8746	38.7966	16.0847
0.2763	10.0	700	0.2954	39.358	22.0903	39.4452	39.5052	16.1017
0.2763	11.0	770	0.2963	38.7295	21.9111	38.8486	38.7888	16.0847
0.2763	12.0	840	0.2950	38.6733	22.0903	38.868	38.8268	16.0847
0.2763	13.0	910	0.2969	39.2337	22.0903	39.3166	39.3686	16.0847
0.2763	14.0	980	0.2943	39.3254	22.2055	39.4837	39.5325	16.0678
0.2694	15.0	1050	0.2939	39.1597	21.9323	39.2799	39.2837	16.0847
0.2694	16.0	1120	0.2942	39.4126	22.0128	39.5265	39.5714	16.0847
0.2694	17.0	1190	0.2971	39.7798	22.728	39.9793	40.0021	16.0847
0.2694	18.0	1260	0.2956	39.701	22.5979	39.8913	39.929	16.0847
0.2694	19.0	1330	0.2945	39.6161	22.2776	39.7607	39.84	16.089
0.2694	20.0	1400	0.2947	39.4039	22.3206	39.5051	39.5487	16.0847
0.2694	21.0	1470	0.2945	39.7001	22.5777	39.8644	39.9219	16.0847
0.2632	22.0	1540	0.2943	39.4609	22.1144	39.5673	39.5833	16.089
0.2632	23.0	1610	0.2946	39.5428	22.3434	39.6705	39.7451	16.0847
0.2632	24.0	1680	0.2946	39.0183	22.0903	39.1424	39.1981	16.0847
0.2632	25.0	1750	0.2955	39.2686	22.4778	39.3774	39.3981	16.0847
0.2632	26.0	1820	0.2955	38.7063	21.936	38.8518	38.8452	16.072
0.2632	27.0	1890	0.2943	39.2686	22.5219	39.3774	39.3981	16.0678
0.2632	28.0	1960	0.2919	39.7635	22.5543	39.9236	40.0061	16.0932
0.258	29.0	2030	0.2911	39.7166	22.5549	39.9268	39.9832	16.0678
0.258	30.0	2100	0.2905	39.4208	22.6958	39.5539	39.5859	16.0763
0.258	31.0	2170	0.2917	39.4279	22.6251	39.575	39.5743	16.0763
0.258	32.0	2240	0.2904	39.6284	22.3963	39.8328	39.9046	16.0678
0.258	33.0	2310	0.2937	39.3461	22.5787	39.5004	39.4931	16.0678
0.258	34.0	2380	0.2897	39.3954	22.5787	39.5232	39.5061	16.0678
0.258	35.0	2450	0.2924	39.4158	22.5787	39.6149	39.5978	16.0678
0.2523	36.0	2520	0.2927	39.5072	22.702	39.6958	39.6845	16.0678
0.2523	37.0	2590	0.2928	39.4158	22.5787	39.6149	39.5978	16.0678
0.2523	38.0	2660	0.2899	39.4397	22.4174	39.5664	39.6141	16.072
0.2523	39.0	2730	0.2917	39.5985	22.5787	39.7005	39.7516	16.0932
0.2523	40.0	2800	0.2920	39.4158	22.5787	39.6149	39.5978	16.0763
0.2523	41.0	2870	0.2898	39.3254	22.6509	39.4715	39.508	16.0678
0.2523	42.0	2940	0.2913	39.4994	22.8155	39.6122	39.5795	16.0847
0.2489	43.0	3010	0.2902	39.1453	22.5787	39.3306	39.2342	16.0847
0.2489	44.0	3080	0.2903	39.778	22.8155	39.856	39.891	16.0932
0.2489	45.0	3150	0.2896	39.3954	22.5787	39.5232	39.5061	16.0678
0.2489	46.0	3220	0.2899	39.659	22.8155	39.7524	39.7486	16.0847
0.2489	47.0	3290	0.2887	39.538	22.8155	39.6348	39.6385	16.0847
0.2489	48.0	3360	0.2905	39.538	22.8155	39.6348	39.6385	16.0678
0.2489	49.0	3430	0.2888	39.6043	22.5446	39.6811	39.6975	16.072
0.2442	50.0	3500	0.2905	39.6579	22.8155	39.7436	39.7544	16.0678
0.2442	51.0	3570	0.2917	39.6978	22.8155	39.7623	39.7429	16.0678
0.2442	52.0	3640	0.2886	39.5685	22.6587	39.6505	39.6604	16.0636
0.2442	53.0	3710	0.2893	39.6489	22.8155	39.7248	39.7425	16.0847
0.2442	54.0	3780	0.2910	39.6489	22.8155	39.7248	39.7425	16.0678
0.2442	55.0	3850	0.2900	39.7014	22.8155	39.7836	39.807	16.072
0.2442	56.0	3920	0.2893	39.7156	22.8155	39.8059	39.7891	16.072
0.2442	57.0	3990	0.2893	39.6579	22.8155	39.7436	39.7544	16.0847
0.2406	58.0	4060	0.2890	39.3975	22.1901	39.4816	39.5241	16.0763
0.2406	59.0	4130	0.2883	39.6046	22.2588	39.7636	39.7933	16.072
0.2406	60.0	4200	0.2895	39.8147	22.8155	39.8815	39.9257	16.0847
0.2406	61.0	4270	0.2900	39.6523	22.6587	39.7435	39.7004	16.072
0.2406	62.0	4340	0.2876	39.4672	22.6587	39.5554	39.533	16.072
0.2406	63.0	4410	0.2872	39.3354	22.4499	39.3997	39.4517	16.0636
0.2406	64.0	4480	0.2898	39.5053	22.1218	39.6356	39.5945	16.0763
0.2379	65.0	4550	0.2897	39.4043	22.4499	39.5828	39.5649	16.072
0.2379	66.0	4620	0.2897	39.7377	22.3954	39.8376	39.8243	16.0763
0.2379	67.0	4690	0.2898	39.5873	22.1218	39.7298	39.6907	16.0763
0.2379	68.0	4760	0.2889	39.5053	22.1218	39.6356	39.5945	16.0763
0.2379	69.0	4830	0.2901	39.5053	22.1218	39.6356	39.5945	16.0763
0.2379	70.0	4900	0.2889	39.6293	22.1218	39.7629	39.7467	16.0763
0.2379	71.0	4970	0.2888	39.5053	22.1218	39.6356	39.5945	16.0763
0.2354	72.0	5040	0.2891	39.5053	22.1218	39.6356	39.5945	16.0763
0.2354	73.0	5110	0.2893	39.5053	22.1218	39.6356	39.5945	16.0847
0.2354	74.0	5180	0.2897	39.5053	22.1218	39.6356	39.5945	16.0763
0.2354	75.0	5250	0.2894	39.6293	22.1218	39.7629	39.7467	16.0847
0.2354	76.0	5320	0.2892	39.6293	22.1218	39.7629	39.7467	16.0847
0.2354	77.0	5390	0.2893	39.5053	22.1218	39.6356	39.5945	16.0847
0.2354	78.0	5460	0.2885	39.6293	22.1218	39.7629	39.7467	16.0847
0.2337	79.0	5530	0.2891	39.6293	22.1218	39.7629	39.7467	16.0763
0.2337	80.0	5600	0.2888	39.5053	22.1218	39.6356	39.5945	16.0763
0.2337	81.0	5670	0.2885	39.5053	22.1218	39.6356	39.5945	16.0763
0.2337	82.0	5740	0.2889	39.5053	22.1218	39.6356	39.5945	16.0763
0.2337	83.0	5810	0.2886	39.5053	22.1218	39.6356	39.5945	16.0763
0.2337	84.0	5880	0.2894	39.5053	22.1218	39.6356	39.5945	16.0763
0.2337	85.0	5950	0.2889	39.5053	22.1218	39.6356	39.5945	16.0763
0.2318	86.0	6020	0.2885	39.5053	22.1218	39.6356	39.5945	16.0763
0.2318	87.0	6090	0.2887	39.5053	22.1218	39.6356	39.5945	16.0763
0.2318	88.0	6160	0.2883	39.5053	22.1218	39.6356	39.5945	16.0763
0.2318	89.0	6230	0.2880	39.4264	22.1218	39.5743	39.5557	16.0763
0.2318	90.0	6300	0.2883	39.7689	22.4239	39.9782	39.9153	16.0805
0.2318	91.0	6370	0.2886	39.7689	22.4239	39.9782	39.9153	16.0805
0.2318	92.0	6440	0.2887	39.8328	22.4239	39.9834	39.9724	16.0805
0.2325	93.0	6510	0.2884	39.9004	22.4239	40.1203	40.0835	16.072
0.2325	94.0	6580	0.2886	39.8328	22.4239	39.9834	39.9724	16.072
0.2325	95.0	6650	0.2890	39.8328	22.4239	39.9834	39.9724	16.0805
0.2325	96.0	6720	0.2889	39.8328	22.4239	39.9834	39.9724	16.0805
0.2325	97.0	6790	0.2889	39.8328	22.4239	39.9834	39.9724	16.0805
0.2325	98.0	6860	0.2889	39.8328	22.4239	39.9834	39.9724	16.0805
0.2325	99.0	6930	0.2889	39.8328	22.4239	39.9834	39.9724	16.0805
0.2303	100.0	7000	0.2889	39.8328	22.4239	39.9834	39.9724	16.0805

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Tokenizers 0.20.3