t5-base_cola_moe_ex19_epochs-3_decoder_all_sparsity10_mare_mlp

This model is a fine-tuned version of t5-base on the glue dataset. It achieves the following results on the evaluation set:

• Loss: 0.6160
• Accuracy: 0.8322

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: 16
• eval_batch_size: 32
• seed: 1
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 20
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5636	0.19	50	0.9030	0.8255
0.5623	0.37	100	0.7397	0.8322
0.571	0.56	150	0.7188	0.8159
0.4997	0.75	200	0.6449	0.8322
0.5069	0.93	250	0.5668	0.8332
0.374	1.12	300	0.6804	0.8245
0.3617	1.31	350	0.6122	0.8313
0.3928	1.5	400	0.5891	0.8274
0.3772	1.68	450	0.6124	0.8245
0.3275	1.87	500	0.5892	0.8255
0.2992	2.06	550	0.6055	0.8255
0.4092	2.24	600	0.6054	0.8293
0.288	2.43	650	0.5972	0.8313
0.3493	2.62	700	0.6449	0.8313
0.2419	2.8	750	0.6198	0.8332
0.3811	2.99	800	0.6252	0.8322

Framework versions

• Transformers 4.34.1
• Pytorch 2.0.1+cu117
• Datasets 2.9.0
• Tokenizers 0.14.1