mt5_deed_sum_1

This model is a fine-tuned version of google/mt5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5951
• Rouge1: 0.0572
• Rouge2: 0.0
• Rougel: 0.0572
• Rougelsum: 0.0572
• Gen Len: 19.0

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5000
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
24.1526	1.0	375	15.7524	0.7082	0.0	0.6842	0.6883	12.0881
11.2879	2.0	750	13.1581	0.7082	0.0	0.6842	0.6883	12.3208
10.5585	3.0	1125	7.4334	0.7082	0.0	0.6842	0.6883	17.1384
2.9668	4.0	1500	6.5357	0.7487	0.0	0.7176	0.7257	19.0
3.7198	5.0	1875	2.1303	0.7487	0.0	0.7176	0.7257	19.0
1.3383	6.0	2250	1.8675	0.7487	0.0	0.7176	0.7257	19.0
3.4973	7.0	2625	1.6299	0.7487	0.0	0.7176	0.7257	18.8994
11.7006	8.0	3000	4.6990	0.7487	0.0	0.7176	0.7257	19.0
0.4529	9.0	3375	1.0729	0.7487	0.0	0.7176	0.7257	19.0
1.2783	10.0	3750	0.9424	0.7487	0.0	0.7176	0.7257	19.0
0.7953	11.0	4125	0.8889	1.0108	0.2013	0.892	0.9014	19.0
0.9359	12.0	4500	0.7996	1.1489	0.3019	0.9952	0.9877	19.0
0.5759	13.0	4875	0.7622	0.1572	0.1144	0.1572	0.1572	19.0
0.1533	14.0	5250	0.7068	0.2144	0.1144	0.1715	0.1715	19.0
0.4524	15.0	5625	0.6760	0.3145	0.2287	0.3145	0.3145	19.0
1.0126	16.0	6000	0.6627	0.0	0.0	0.0	0.0	19.0
0.1065	17.0	6375	0.6391	0.3145	0.2287	0.3145	0.3145	19.0
0.2096	18.0	6750	0.6419	0.2144	0.1144	0.2144	0.2144	19.0
0.5649	19.0	7125	0.6261	0.1572	0.1144	0.1572	0.1572	19.0
0.125	20.0	7500	0.6139	0.1572	0.1144	0.1572	0.1572	19.0
0.5511	21.0	7875	0.6057	0.3145	0.2287	0.3145	0.3145	19.0
0.0759	22.0	8250	0.6029	0.3145	0.2287	0.3145	0.3145	19.0
0.3491	23.0	8625	0.5995	0.3145	0.2287	0.3145	0.3145	19.0
0.3735	24.0	9000	0.5936	0.1572	0.1144	0.1572	0.1572	19.0
0.3451	25.0	9375	0.5951	0.0572	0.0	0.0572	0.0572	19.0

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0.dev20230811+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2