e5_Sentiment_General_v2
This model is a fine-tuned version of intfloat/multilingual-e5-large-instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4553
- Accuracy: 0.9016
- F1: 0.9021
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.4454 | 1.0 | 328 | 0.3933 | 0.8716 | 0.8738 |
| 0.3303 | 2.0 | 656 | 0.3731 | 0.8808 | 0.8832 |
| 0.2655 | 3.0 | 984 | 0.4259 | 0.9000 | 0.9012 |
| 0.1864 | 4.0 | 1312 | 0.4484 | 0.9008 | 0.9014 |
| 0.1207 | 5.0 | 1640 | 0.4553 | 0.9016 | 0.9021 |
Framework versions
- Transformers 4.53.2
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.21.2