|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: FacebookAI/xlm-roberta-base |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- precision |
|
|
- recall |
|
|
- accuracy |
|
|
model-index: |
|
|
- name: finerweb-multilabel-classifier-xlmr-4o |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
|
|
# finerweb-multilabel-classifier-xlmr-4o |
|
|
|
|
|
This model is a fine-tuned version of [FacebookAI/xlm-roberta-base](https://huggingface.co/FacebookAI/xlm-roberta-base) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.3645 |
|
|
- Precision: 0.5930 |
|
|
- Recall: 0.4813 |
|
|
- F1 Macro: 0.5091 |
|
|
- Accuracy: 0.6488 |
|
|
|
|
|
## 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: 64 |
|
|
- eval_batch_size: 128 |
|
|
- seed: 0 |
|
|
- 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 |
|
|
- num_epochs: 20 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 Macro | Accuracy | |
|
|
|:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:--------:|:--------:| |
|
|
| No log | 0 | 0 | 8.1215 | 0.0002 | 0.2 | 0.0004 | 0.0010 | |
|
|
| 0.2936 | 0.7812 | 1000 | 0.2877 | 0.5095 | 0.4544 | 0.4485 | 0.6647 | |
|
|
| 0.2516 | 1.5625 | 2000 | 0.3197 | 0.5299 | 0.3886 | 0.4136 | 0.6293 | |
|
|
| 0.1737 | 2.3438 | 3000 | 0.2922 | 0.5089 | 0.4296 | 0.4518 | 0.6516 | |
|
|
| 0.0996 | 3.125 | 4000 | 0.3233 | 0.5012 | 0.4379 | 0.4585 | 0.6369 | |
|
|
| 0.1051 | 3.9062 | 5000 | 0.3108 | 0.5042 | 0.4401 | 0.4609 | 0.6496 | |
|
|
| 0.0601 | 4.6875 | 6000 | 0.3501 | 0.4840 | 0.4501 | 0.4614 | 0.6411 | |
|
|
| 0.0588 | 5.4688 | 7000 | 0.3554 | 0.4758 | 0.4585 | 0.4658 | 0.6327 | |
|
|
| 0.0385 | 6.25 | 8000 | 0.3527 | 0.4853 | 0.4518 | 0.4647 | 0.6331 | |
|
|
| 0.0312 | 7.0312 | 9000 | 0.3475 | 0.4912 | 0.4580 | 0.4714 | 0.6415 | |
|
|
| 0.0314 | 7.8125 | 10000 | 0.3442 | 0.4983 | 0.4557 | 0.4679 | 0.6551 | |
|
|
| 0.0211 | 8.5938 | 11000 | 0.3543 | 0.4823 | 0.4740 | 0.4769 | 0.6449 | |
|
|
| 0.0193 | 9.375 | 12000 | 0.3585 | 0.4816 | 0.4621 | 0.4697 | 0.6438 | |
|
|
| 0.0208 | 10.1562 | 13000 | 0.3730 | 0.4825 | 0.4588 | 0.4659 | 0.6207 | |
|
|
| 0.0203 | 10.9375 | 14000 | 0.3578 | 0.4903 | 0.4748 | 0.4818 | 0.6468 | |
|
|
| 0.0154 | 11.7188 | 15000 | 0.3513 | 0.4994 | 0.4591 | 0.4744 | 0.6557 | |
|
|
| 0.0126 | 12.5 | 16000 | 0.3623 | 0.4920 | 0.4488 | 0.4649 | 0.6460 | |
|
|
| 0.0087 | 13.2812 | 17000 | 0.3632 | 0.4940 | 0.4512 | 0.4675 | 0.6449 | |
|
|
| 0.0091 | 14.0625 | 18000 | 0.3518 | 0.6976 | 0.4793 | 0.5121 | 0.6533 | |
|
|
| 0.009 | 14.8438 | 19000 | 0.3569 | 0.5918 | 0.4886 | 0.5128 | 0.6527 | |
|
|
| 0.0056 | 15.625 | 20000 | 0.3672 | 0.5532 | 0.4882 | 0.5081 | 0.6457 | |
|
|
| 0.0045 | 16.4062 | 21000 | 0.3655 | 0.5391 | 0.4870 | 0.5060 | 0.6460 | |
|
|
| 0.0035 | 17.1875 | 22000 | 0.3646 | 0.4955 | 0.4634 | 0.4765 | 0.6489 | |
|
|
| 0.0032 | 17.9688 | 23000 | 0.3631 | 0.6942 | 0.4841 | 0.5150 | 0.6503 | |
|
|
| 0.0035 | 18.75 | 24000 | 0.3625 | 0.5986 | 0.4805 | 0.5103 | 0.6504 | |
|
|
| 0.0019 | 19.5312 | 25000 | 0.3645 | 0.5930 | 0.4813 | 0.5091 | 0.6488 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.49.0 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 3.3.2 |
|
|
- Tokenizers 0.21.1 |
|
|
|
