t5_summarization_results

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

• Loss: 1.1410
• Rouge1: 73.1291
• Rouge2: 53.3202
• Rougel: 70.4631
• Rougelsum: 70.571
• Gen Len: 54.8889

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: 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_steps: 10
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.4356	0.7273	2	1.9470	66.0971	41.4032	60.6429	60.5259	57.1111
2.7317	1.3636	4	1.9059	65.0871	41.1833	59.5891	59.4131	80.0
2.0131	2.0	6	1.8426	65.1386	41.2334	59.6397	59.3273	80.0
2.3027	2.7273	8	1.7613	63.4406	39.9188	57.8757	57.2837	76.7778
2.0473	3.3636	10	1.6736	65.013	41.1626	59.4678	59.1496	80.4444
1.9981	4.0	12	1.5863	63.5392	40.5478	58.1394	57.6228	76.7778
1.8935	4.7273	14	1.5155	63.5392	40.5478	58.1394	57.6228	76.7778
1.8547	5.3636	16	1.4578	65.9514	41.0767	59.6275	59.2967	58.5556
1.6267	6.0	18	1.4100	70.7089	44.4184	63.9004	63.7558	63.3333
1.598	6.7273	20	1.3699	71.1296	43.94	64.0466	63.9285	63.7778
1.4815	7.3636	22	1.3349	72.1911	44.7003	65.3024	65.3564	61.1111
1.4331	8.0	24	1.3055	72.1038	45.5051	65.4691	65.4648	56.6667
1.513	8.7273	26	1.2809	73.4532	46.3707	66.7577	66.8106	56.7778
1.6751	9.3636	28	1.2609	72.0386	46.1991	65.5128	65.5993	56.1111
1.4712	10.0	30	1.2443	71.2442	48.419	66.548	66.5424	55.6667
1.4576	10.7273	32	1.2290	71.054	48.5114	66.8665	66.8427	55.6667
1.6172	11.3636	34	1.2153	72.9291	49.7383	68.7427	68.7354	54.8889
1.1523	12.0	36	1.2030	73.6441	51.3112	70.0868	70.1427	54.4444
1.5492	12.7273	38	1.1912	71.2174	50.6641	68.6358	68.5363	54.0
1.388	13.3636	40	1.1812	71.3115	50.9578	68.9737	68.9281	54.6667
1.3034	14.0	42	1.1728	69.7521	49.8104	67.6361	67.5158	55.1111
1.1849	14.7273	44	1.1663	69.7521	49.8104	67.6361	67.5158	55.1111
1.2925	15.3636	46	1.1601	72.49	52.0082	69.8034	69.9001	55.1111
1.4154	16.0	48	1.1549	72.9973	53.2788	70.283	70.3955	54.8889
1.4613	16.7273	50	1.1508	72.9973	53.2788	70.283	70.3955	54.8889
1.0993	17.3636	52	1.1474	72.9973	53.2788	70.283	70.3955	54.8889
1.6538	18.0	54	1.1446	73.1291	53.3202	70.4631	70.571	54.8889
1.203	18.7273	56	1.1427	73.1291	53.3202	70.4631	70.571	54.8889
1.1785	19.3636	58	1.1416	73.1291	53.3202	70.4631	70.571	54.8889
1.1342	20.0	60	1.1410	73.1291	53.3202	70.4631	70.571	54.8889

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1