my_awesome_billsum_model_15

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: 0.4391
• Rouge1: 0.9758
• Rouge2: 0.8793
• Rougel: 0.9297
• Rougelsum: 0.9308
• Gen Len: 5.3958

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	2.0226	0.3972	0.2663	0.3661	0.3661	18.125
No log	2.0	24	1.4195	0.4329	0.299	0.3963	0.3964	16.75
No log	3.0	36	0.8497	0.5804	0.4564	0.5365	0.5379	13.0417
No log	4.0	48	0.5856	0.9414	0.7914	0.8515	0.8524	5.375
No log	5.0	60	0.5442	0.9541	0.8097	0.8646	0.867	5.0417
No log	6.0	72	0.5065	0.9561	0.8118	0.8655	0.869	5.0417
No log	7.0	84	0.4694	0.9591	0.8236	0.8744	0.8774	5.0625
No log	8.0	96	0.4401	0.9568	0.8153	0.8677	0.8705	5.0833
No log	9.0	108	0.4206	0.9692	0.8608	0.8991	0.9008	5.1875
No log	10.0	120	0.4085	0.9722	0.872	0.9077	0.9091	5.2083
No log	11.0	132	0.4067	0.9676	0.8617	0.9048	0.9053	5.2917
No log	12.0	144	0.4094	0.9653	0.8535	0.8938	0.8936	5.3125
No log	13.0	156	0.4080	0.9676	0.8507	0.8963	0.8989	5.2917
No log	14.0	168	0.4005	0.9676	0.8507	0.8963	0.8989	5.2917
No log	15.0	180	0.3993	0.9676	0.8507	0.8963	0.8989	5.2917
No log	16.0	192	0.3921	0.9676	0.8507	0.8963	0.8989	5.2917
No log	17.0	204	0.3880	0.9676	0.8507	0.8963	0.8989	5.2917
No log	18.0	216	0.3879	0.9676	0.8507	0.8963	0.8989	5.2917
No log	19.0	228	0.3900	0.9707	0.8643	0.9059	0.9078	5.3125
No log	20.0	240	0.3914	0.9737	0.8777	0.9163	0.9158	5.3333
No log	21.0	252	0.3933	0.9695	0.8551	0.9042	0.9046	5.3542
No log	22.0	264	0.3938	0.9695	0.8551	0.9042	0.9046	5.3542
No log	23.0	276	0.3958	0.9695	0.8551	0.9042	0.9046	5.3542
No log	24.0	288	0.3993	0.9695	0.8551	0.9042	0.9046	5.3542
No log	25.0	300	0.3957	0.9695	0.8551	0.9042	0.9046	5.3542
No log	26.0	312	0.3934	0.9695	0.8551	0.9042	0.9046	5.3542
No log	27.0	324	0.3963	0.9695	0.8551	0.9042	0.9046	5.3542
No log	28.0	336	0.3977	0.9695	0.8551	0.9042	0.9046	5.3542
No log	29.0	348	0.3951	0.9695	0.8551	0.9042	0.9046	5.3542
No log	30.0	360	0.3966	0.9661	0.8551	0.9051	0.9051	5.3333
No log	31.0	372	0.3962	0.9695	0.8551	0.9042	0.9046	5.3542
No log	32.0	384	0.3950	0.9695	0.8551	0.9042	0.9046	5.3542
No log	33.0	396	0.3859	0.9695	0.8551	0.9042	0.9046	5.3542
No log	34.0	408	0.3869	0.9668	0.8534	0.9018	0.9026	5.375
No log	35.0	420	0.3871	0.9668	0.8534	0.9018	0.9026	5.375
No log	36.0	432	0.3823	0.9668	0.8534	0.9018	0.9026	5.375
No log	37.0	444	0.3869	0.9698	0.867	0.9115	0.9114	5.3542
No log	38.0	456	0.3934	0.9668	0.8534	0.9018	0.9026	5.375
No log	39.0	468	0.3960	0.9668	0.8534	0.9018	0.9026	5.375
No log	40.0	480	0.3977	0.9698	0.867	0.9115	0.9114	5.3542
No log	41.0	492	0.3991	0.966	0.8599	0.912	0.9133	5.375
0.4754	42.0	504	0.4013	0.966	0.8599	0.912	0.9133	5.375
0.4754	43.0	516	0.4082	0.966	0.8599	0.912	0.9133	5.375
0.4754	44.0	528	0.4055	0.9729	0.8664	0.9205	0.9216	5.4167
0.4754	45.0	540	0.4017	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	46.0	552	0.3980	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	47.0	564	0.3990	0.9691	0.8755	0.9193	0.9198	5.3125
0.4754	48.0	576	0.4030	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	49.0	588	0.4094	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	50.0	600	0.4092	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	51.0	612	0.4078	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	52.0	624	0.4083	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	53.0	636	0.4083	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	54.0	648	0.4041	0.9691	0.8755	0.9193	0.9198	5.3125
0.4754	55.0	660	0.4090	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	56.0	672	0.4117	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	57.0	684	0.4185	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	58.0	696	0.4219	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	59.0	708	0.4233	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	60.0	720	0.4202	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	61.0	732	0.4225	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	62.0	744	0.4291	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	63.0	756	0.4311	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	64.0	768	0.4293	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	65.0	780	0.4337	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	66.0	792	0.4346	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	67.0	804	0.4354	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	68.0	816	0.4364	0.9695	0.873	0.9214	0.9227	5.3542
0.4754	69.0	828	0.4380	0.9695	0.8877	0.9263	0.9271	5.3542
0.4754	70.0	840	0.4375	0.9758	0.8933	0.935	0.936	5.3958
0.4754	71.0	852	0.4397	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	72.0	864	0.4382	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	73.0	876	0.4386	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	74.0	888	0.4387	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	75.0	900	0.4378	0.9758	0.8933	0.935	0.936	5.3958
0.4754	76.0	912	0.4394	0.9758	0.8933	0.935	0.936	5.3958
0.4754	77.0	924	0.4409	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	78.0	936	0.4429	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	79.0	948	0.4434	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	80.0	960	0.4421	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	81.0	972	0.4405	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	82.0	984	0.4407	0.9758	0.8793	0.9297	0.9308	5.3958
0.4754	83.0	996	0.4396	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	84.0	1008	0.4415	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	85.0	1020	0.4410	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	86.0	1032	0.4401	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	87.0	1044	0.4381	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	88.0	1056	0.4370	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	89.0	1068	0.4366	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	90.0	1080	0.4354	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	91.0	1092	0.4355	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	92.0	1104	0.4359	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	93.0	1116	0.4374	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	94.0	1128	0.4372	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	95.0	1140	0.4376	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	96.0	1152	0.4378	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	97.0	1164	0.4386	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	98.0	1176	0.4389	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	99.0	1188	0.4389	0.9758	0.8793	0.9297	0.9308	5.3958
0.1151	100.0	1200	0.4391	0.9758	0.8793	0.9297	0.9308	5.3958

Framework versions

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