e5-base-v2-sentiment-twitter

This model is a fine-tuned version of intfloat/e5-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6678
• Accuracy: 0.7186
• F1: 0.7186
• Precision: 0.7219
• Recall: 0.7186

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.6842	0.1754	500	0.6403	0.7195	0.7209	0.7232	0.7195
0.6395	0.3508	1000	0.6110	0.7215	0.7254	0.7374	0.7215
0.6188	0.5261	1500	0.6028	0.733	0.7360	0.7442	0.733
0.6291	0.7015	2000	0.5912	0.738	0.7338	0.7403	0.738
0.6005	0.8769	2500	0.5705	0.752	0.7534	0.7572	0.752
0.3942	1.0523	3000	0.6278	0.747	0.7469	0.7525	0.747
0.4603	1.2276	3500	0.6185	0.75	0.7509	0.7536	0.75
0.4579	1.4030	4000	0.6348	0.751	0.7491	0.7526	0.751
0.4264	1.5784	4500	0.6129	0.757	0.7573	0.7579	0.757
0.4196	1.7538	5000	0.6196	0.7585	0.7582	0.7582	0.7585
0.4193	1.9291	5500	0.6159	0.7625	0.7611	0.7615	0.7625

Framework versions

• Transformers 4.55.4
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4