longt5_xl_sfd_bp_10

This model is a fine-tuned version of google/long-t5-tglobal-xl on the learn3r/summ_screen_fd_bp dataset. It achieves the following results on the evaluation set:

• Loss: 2.8921

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss
2.3973	0.97	14	1.9027
1.9188	1.95	28	1.6941
1.4297	2.99	43	1.5011
1.2759	3.97	57	1.5048
1.1421	4.94	71	1.5463
0.9605	5.98	86	1.6270
0.8082	6.96	100	1.7646
0.664	8.0	115	1.7878
0.5471	8.97	129	1.9500
0.4349	9.95	143	1.9657
0.4338	10.99	158	2.1351
0.2887	11.97	172	2.1166
0.2753	12.94	186	2.4357
0.2114	13.98	201	2.5789
0.1805	14.96	215	2.6075
0.1543	16.0	230	2.5597
0.5166	16.97	244	2.5067
0.1117	17.95	258	2.8087
0.0895	18.99	273	2.7578
0.0779	19.48	280	2.8921

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0