ghc-google-t5-v1_1-large-inter_model-dataset-frequency-human_annots_str

This model is a fine-tuned version of google/t5-v1_1-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4160

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss
2.5863	1.0	345	2.2862
1.9673	2.0	690	2.0705
1.7865	3.0	1035	1.8048
0.0714	4.0	1380	0.0459
0.0618	5.0	1725	0.0456
0.0596	6.0	2070	0.0476
0.0532	7.0	2415	0.0438
0.0503	8.0	2760	0.0405
0.048	9.0	3105	0.0377
0.0462	10.0	3450	0.0455
0.036	11.0	3795	0.0358
0.0447	12.0	4140	0.0355
0.0416	13.0	4485	0.0351
0.0413	14.0	4830	0.0331
0.0409	15.0	5175	0.0320
0.0411	16.0	5520	0.0333
0.0363	17.0	5865	0.0322
0.0378	18.0	6210	0.0329
0.0345	19.0	6555	0.0312
0.0328	20.0	6900	0.0311
0.0392	21.0	7245	0.0303
0.0392	22.0	7590	0.0296
0.0353	23.0	7935	0.0300
0.0331	24.0	8280	0.0299
0.0306	25.0	8625	0.0290
0.0313	26.0	8970	0.0294
0.0303	27.0	9315	0.0296
0.0378	28.0	9660	0.0292
0.0358	29.0	10005	0.0292
0.0328	30.0	10350	0.0292

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.14.1