my_awesome_billsum_model_62

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

• Loss: 0.7970
• Rouge1: 0.9571
• Rouge2: 0.8259
• Rougel: 0.8928
• Rougelsum: 0.8902
• Gen Len: 5.0208

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: 16
• eval_batch_size: 16
• 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	1.0	12	2.4286	0.3894	0.2336	0.3514	0.3508	17.8542
No log	2.0	24	1.8139	0.4266	0.2737	0.389	0.3886	16.4167
No log	3.0	36	1.2636	0.6493	0.4505	0.568	0.5646	11.1042
No log	4.0	48	1.0763	0.9258	0.7101	0.8078	0.8059	4.9792
No log	5.0	60	1.0843	0.935	0.7341	0.8244	0.8199	5.0833
No log	6.0	72	1.0524	0.9404	0.7398	0.8318	0.8271	4.7917
No log	7.0	84	0.9935	0.9404	0.7398	0.8318	0.8271	4.7917
No log	8.0	96	0.9337	0.9461	0.7441	0.8277	0.827	4.875
No log	9.0	108	0.9054	0.9491	0.7772	0.8475	0.8461	4.8958
No log	10.0	120	0.8916	0.9491	0.7772	0.8475	0.8461	4.8958
No log	11.0	132	0.8979	0.9514	0.7797	0.8496	0.8483	4.9375
No log	12.0	144	0.8762	0.9514	0.7797	0.8496	0.8483	4.9375
No log	13.0	156	0.8374	0.9514	0.7797	0.8496	0.8483	4.9375
No log	14.0	168	0.8129	0.9496	0.7903	0.8673	0.8652	5.0
No log	15.0	180	0.7959	0.9496	0.7903	0.8673	0.8652	5.0
No log	16.0	192	0.7882	0.9496	0.7903	0.8673	0.8652	5.0
No log	17.0	204	0.7801	0.9516	0.791	0.8642	0.8611	4.9792
No log	18.0	216	0.7644	0.9516	0.791	0.8642	0.8611	4.9792
No log	19.0	228	0.7450	0.9496	0.7903	0.8673	0.8652	5.0
No log	20.0	240	0.7485	0.9474	0.7847	0.8589	0.8566	4.9583
No log	21.0	252	0.7483	0.9498	0.7857	0.8551	0.8537	4.9375
No log	22.0	264	0.7495	0.9452	0.7942	0.8701	0.8681	4.9792
No log	23.0	276	0.7544	0.9476	0.7955	0.866	0.8646	4.9583
No log	24.0	288	0.7588	0.9498	0.7971	0.8623	0.8598	4.9375
No log	25.0	300	0.7542	0.9523	0.8027	0.87	0.8689	4.9792
No log	26.0	312	0.7427	0.9523	0.7919	0.8629	0.8615	4.9792
No log	27.0	324	0.7295	0.9463	0.7886	0.8647	0.8631	5.0208
No log	28.0	336	0.7257	0.9463	0.7886	0.8647	0.8631	5.0208
No log	29.0	348	0.7276	0.9498	0.8014	0.8738	0.8727	5.0417
No log	30.0	360	0.7367	0.9498	0.8014	0.8738	0.8727	5.0417
No log	31.0	372	0.7455	0.9549	0.8155	0.8804	0.8771	5.0
No log	32.0	384	0.7482	0.9549	0.8155	0.8804	0.8771	5.0
No log	33.0	396	0.7448	0.9522	0.8028	0.8698	0.8691	5.0208
No log	34.0	408	0.7516	0.9491	0.7899	0.8609	0.8601	5.0
No log	35.0	420	0.7536	0.9491	0.7899	0.8609	0.8601	5.0
No log	36.0	432	0.7522	0.9522	0.8028	0.8698	0.8691	5.0208
No log	37.0	444	0.7485	0.9522	0.8028	0.8698	0.8691	5.0208
No log	38.0	456	0.7476	0.9522	0.7956	0.8698	0.8691	5.0208
No log	39.0	468	0.7528	0.9522	0.7956	0.8698	0.8691	5.0208
No log	40.0	480	0.7573	0.9522	0.7956	0.8698	0.8691	5.0208
No log	41.0	492	0.7593	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	42.0	504	0.7629	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	43.0	516	0.7512	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	44.0	528	0.7405	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	45.0	540	0.7307	0.955	0.8251	0.8969	0.894	5.0417
0.4192	46.0	552	0.7344	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	47.0	564	0.7373	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	48.0	576	0.7474	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	49.0	588	0.7551	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	50.0	600	0.7698	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	51.0	612	0.7650	0.9542	0.8037	0.8773	0.8764	5.0
0.4192	52.0	624	0.7509	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	53.0	636	0.7529	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	54.0	648	0.7593	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	55.0	660	0.7594	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	56.0	672	0.7623	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	57.0	684	0.7701	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	58.0	696	0.7710	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	59.0	708	0.7684	0.959	0.8279	0.8891	0.8867	5.0
0.4192	60.0	720	0.7661	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	61.0	732	0.7649	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	62.0	744	0.7722	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	63.0	756	0.7689	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	64.0	768	0.7618	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	65.0	780	0.7609	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	66.0	792	0.7674	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	67.0	804	0.7722	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	68.0	816	0.7726	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	69.0	828	0.7724	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	70.0	840	0.7750	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	71.0	852	0.7745	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	72.0	864	0.7756	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	73.0	876	0.7798	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	74.0	888	0.7895	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	75.0	900	0.7929	0.959	0.8279	0.8891	0.8867	5.0
0.4192	76.0	912	0.7903	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	77.0	924	0.7869	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	78.0	936	0.7883	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	79.0	948	0.7888	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	80.0	960	0.7918	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	81.0	972	0.7921	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	82.0	984	0.7921	0.9571	0.8259	0.8928	0.8902	5.0208
0.4192	83.0	996	0.7945	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	84.0	1008	0.7962	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	85.0	1020	0.7955	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	86.0	1032	0.7977	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	87.0	1044	0.7991	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	88.0	1056	0.7986	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	89.0	1068	0.7989	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	90.0	1080	0.7995	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	91.0	1092	0.8005	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	92.0	1104	0.7990	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	93.0	1116	0.7980	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	94.0	1128	0.7978	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	95.0	1140	0.7972	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	96.0	1152	0.7966	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	97.0	1164	0.7961	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	98.0	1176	0.7966	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	99.0	1188	0.7972	0.9571	0.8259	0.8928	0.8902	5.0208
0.0933	100.0	1200	0.7970	0.9571	0.8259	0.8928	0.8902	5.0208

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1