Sentiment-google-t5-v1_1-large-intra_model-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: 1.5938

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: 128
• eval_batch_size: 128
• 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
20.8407	1.0	44	24.0718
19.0527	2.0	88	19.4828
13.4842	3.0	132	11.5393
10.8626	4.0	176	10.8995
10.3562	5.0	220	10.7412
10.008	6.0	264	10.5271
9.8519	7.0	308	10.3934
9.6414	8.0	352	10.0350
8.9978	9.0	396	9.4410
8.6735	10.0	440	9.0569
8.3986	11.0	484	8.8689
8.2999	12.0	528	8.7266
1.7304	13.0	572	1.1034
1.165	14.0	616	1.0495
1.0776	15.0	660	1.0454
1.0862	16.0	704	1.0384
1.0628	17.0	748	1.0318
1.0547	18.0	792	1.0329
1.0513	19.0	836	1.0381
1.0389	20.0	880	1.0247
1.0381	21.0	924	1.0231
1.056	22.0	968	1.0160
1.0508	23.0	1012	1.0171
1.0514	24.0	1056	1.0143
1.0373	25.0	1100	1.0128
1.0295	26.0	1144	1.0129
1.0178	27.0	1188	1.0110
1.0216	28.0	1232	1.0056
1.0355	29.0	1276	1.0084
1.0276	30.0	1320	1.0017
1.0066	31.0	1364	1.0080
1.0107	32.0	1408	1.0044
1.0165	33.0	1452	1.0019

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.6.1
• Tokenizers 0.14.1