---
license: apache-2.0
base_model: allenai/longformer-base-4096
tags:
- generated_from_trainer
datasets:
- essays_su_g
metrics:
- accuracy
model-index:
- name: longformer-sep_tok
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: essays_su_g
      type: essays_su_g
      config: sep_tok
      split: test
      args: sep_tok
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.8870016984713758
---

<!-- 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. -->

# longformer-sep_tok

This model is a fine-tuned version of [allenai/longformer-base-4096](https://huggingface.co/allenai/longformer-base-4096) on the essays_su_g dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2615
- Claim: {'precision': 0.6071779744346116, 'recall': 0.5809031044214488, 'f1-score': 0.5937500000000001, 'support': 4252.0}
- Majorclaim: {'precision': 0.8395117540687161, 'recall': 0.8510540788267644, 'f1-score': 0.8452435138825672, 'support': 2182.0}
- O: {'precision': 1.0, 'recall': 0.9992978144474678, 'f1-score': 0.9996487839143033, 'support': 11393.0}
- Premise: {'precision': 0.8835139944992719, 'recall': 0.8952459016393443, 'f1-score': 0.8893412588551421, 'support': 12200.0}
- Accuracy: 0.8870
- Macro avg: {'precision': 0.83255093075065, 'recall': 0.8316252248337562, 'f1-score': 0.8319958891630032, 'support': 30027.0}
- Weighted avg: {'precision': 0.8853833592288615, 'recall': 0.8870016984713758, 'f1-score': 0.8861327572005248, 'support': 30027.0}

## 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Claim                                                                                                              | Majorclaim                                                                                                         | O                                                                                                                  | Premise                                                                                                             | Accuracy | Macro avg                                                                                                           | Weighted avg                                                                                                        |
|:-------------:|:-----:|:----:|:---------------:|:------------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------------------:|:--------:|:-------------------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------------------:|
| No log        | 1.0   | 41   | 0.3496          | {'precision': 0.514827018121911, 'recall': 0.2939793038570085, 'f1-score': 0.37425149700598803, 'support': 4252.0} | {'precision': 0.6678657074340527, 'recall': 0.7658111824014665, 'f1-score': 0.7134927412467975, 'support': 2182.0} | {'precision': 0.996649620878152, 'recall': 0.9921881857280787, 'f1-score': 0.9944138992742467, 'support': 11393.0} | {'precision': 0.826608505997819, 'recall': 0.9319672131147541, 'f1-score': 0.8761317665189751, 'support': 12200.0}  | 0.8524   | {'precision': 0.7514877131079837, 'recall': 0.7459864712753269, 'f1-score': 0.7395724760115018, 'support': 30027.0} | {'precision': 0.8354407819133993, 'recall': 0.8523995071102675, 'f1-score': 0.8381231435918661, 'support': 30027.0} |
| No log        | 2.0   | 82   | 0.2859          | {'precision': 0.5445935280189423, 'recall': 0.486829727187206, 'f1-score': 0.514094126412517, 'support': 4252.0}   | {'precision': 0.8543130990415335, 'recall': 0.6127406049495875, 'f1-score': 0.7136375767280491, 'support': 2182.0} | {'precision': 1.0, 'recall': 0.997366804178004, 'f1-score': 0.9986816663737036, 'support': 11393.0}                | {'precision': 0.851030230109791, 'recall': 0.9276229508196722, 'f1-score': 0.8876774648992078, 'support': 12200.0}  | 0.8688   | {'precision': 0.8124842142925667, 'recall': 0.7561400217836175, 'f1-score': 0.7785227086033695, 'support': 30027.0} | {'precision': 0.8643984304320985, 'recall': 0.8687847603823226, 'f1-score': 0.8642465352746717, 'support': 30027.0} |
| No log        | 3.0   | 123  | 0.2615          | {'precision': 0.617028164454516, 'recall': 0.4482596425211665, 'f1-score': 0.5192753030922217, 'support': 4252.0}  | {'precision': 0.8376436781609196, 'recall': 0.8015582034830431, 'f1-score': 0.8192037470725996, 'support': 2182.0} | {'precision': 1.0, 'recall': 0.9988589484771351, 'f1-score': 0.9994291485531112, 'support': 11393.0}               | {'precision': 0.8495916852264291, 'recall': 0.9380327868852459, 'f1-score': 0.8916244643552784, 'support': 12200.0} | 0.8818   | {'precision': 0.8260658819604662, 'recall': 0.7966773953416476, 'f1-score': 0.8073831657683027, 'support': 30027.0} | {'precision': 0.872859786884143, 'recall': 0.8818396776234723, 'f1-score': 0.874538779080845, 'support': 30027.0}   |
| No log        | 4.0   | 164  | 0.2591          | {'precision': 0.6161943319838057, 'recall': 0.5369238005644402, 'f1-score': 0.5738343596832978, 'support': 4252.0} | {'precision': 0.860332541567696, 'recall': 0.8299725022914757, 'f1-score': 0.8448798693725216, 'support': 2182.0}  | {'precision': 1.0, 'recall': 0.9992100412534012, 'f1-score': 0.9996048645563507, 'support': 11393.0}               | {'precision': 0.8684641159510637, 'recall': 0.9135245901639344, 'f1-score': 0.8904246394758917, 'support': 12200.0} | 0.8866   | {'precision': 0.8362477473756413, 'recall': 0.8199077335683129, 'f1-score': 0.8271859332720154, 'support': 30027.0} | {'precision': 0.8820583514802954, 'recall': 0.8866353615079762, 'f1-score': 0.8837096744876479, 'support': 30027.0} |
| No log        | 5.0   | 205  | 0.2615          | {'precision': 0.6071779744346116, 'recall': 0.5809031044214488, 'f1-score': 0.5937500000000001, 'support': 4252.0} | {'precision': 0.8395117540687161, 'recall': 0.8510540788267644, 'f1-score': 0.8452435138825672, 'support': 2182.0} | {'precision': 1.0, 'recall': 0.9992978144474678, 'f1-score': 0.9996487839143033, 'support': 11393.0}               | {'precision': 0.8835139944992719, 'recall': 0.8952459016393443, 'f1-score': 0.8893412588551421, 'support': 12200.0} | 0.8870   | {'precision': 0.83255093075065, 'recall': 0.8316252248337562, 'f1-score': 0.8319958891630032, 'support': 30027.0}   | {'precision': 0.8853833592288615, 'recall': 0.8870016984713758, 'f1-score': 0.8861327572005248, 'support': 30027.0} |


### Framework versions

- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2