AlexWang99/byt5_add_10k

byt5_add

This model is a fine-tuned version of google/byt5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1606

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: 800
• eval_batch_size: 800
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	13	4.2259
No log	2.0	26	2.4178
No log	3.0	39	1.9256
No log	4.0	52	1.7310
No log	5.0	65	1.6577
No log	6.0	78	1.6385
No log	7.0	91	1.6110
No log	8.0	104	1.5811
No log	9.0	117	1.5237
No log	10.0	130	1.4809
No log	11.0	143	1.4378
No log	12.0	156	1.3976
No log	13.0	169	1.3462
No log	14.0	182	1.2587
No log	15.0	195	1.2260
No log	16.0	208	1.1018
No log	17.0	221	1.0273
No log	18.0	234	0.9436
No log	19.0	247	0.8007
No log	20.0	260	0.6919
No log	21.0	273	0.6201
No log	22.0	286	0.5486
No log	23.0	299	0.4804
No log	24.0	312	0.4080
No log	25.0	325	0.3861
No log	26.0	338	0.3477
No log	27.0	351	0.3181
No log	28.0	364	0.2921
No log	29.0	377	0.2832
No log	30.0	390	0.2693
No log	31.0	403	0.2469
No log	32.0	416	0.2453
No log	33.0	429	0.2313
No log	34.0	442	0.2134
No log	35.0	455	0.2139
No log	36.0	468	0.2088
No log	37.0	481	0.2007
No log	38.0	494	0.1960
1.3	39.0	507	0.1830
1.3	40.0	520	0.1782
1.3	41.0	533	0.1746
1.3	42.0	546	0.1741
1.3	43.0	559	0.1708
1.3	44.0	572	0.1668
1.3	45.0	585	0.1650
1.3	46.0	598	0.1651
1.3	47.0	611	0.1629
1.3	48.0	624	0.1627
1.3	49.0	637	0.1610
1.3	50.0	650	0.1606

Framework versions

• Transformers 4.35.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2