---
license: mit
language:
- en
base_model:
- distilbert/distilbert-base-uncased
pipeline_tag: text-classification
library_name: transformers
tags:
- moderation
- filtering
- offensive
- clean
datasets:
- akaruineko/fantastic-offensive
model-index:
- name: akaruineko/ftan-2.5
  results:
  - task:
      type: text-classification
      id: overall
    dataset:
      type: akaruineko/ftanch
      name: FTANch
      split: test
    metrics:
    - type: acc
      value: 0.8366
      name: Accuracy
    - type: p
      value: 0.7999
      name: Precision
    - type: r
      value: 0.9616
      name: Recall
    - type: f1
      value: 0.8733
      name: F1-Score
  - task:
      type: text-classification
      id: test
    dataset:
      type: akaruineko/ftanch
      name: FTANch
      split: test
    metrics:
    - type: acc
      value: 0.8007
      name: Accuracy
    - type: p
      value: 0.7315
      name: Precision
    - type: r
      value: 0.9499
      name: Recall
    - type: f1
      value: 0.8266
      name: F1-Score
  - task:
      type: text-classification
      id: test_obfuscated
    dataset:
      type: akaruineko/ftanch
      name: FTANch
      split: test
    metrics:
    - type: acc
      value: 0.9134
      name: Accuracy
    - type: p
      value: 0.9152
      name: Precision
    - type: r
      value: 0.9778
      name: Recall
    - type: f1
      value: 0.9455
      name: F1-Score
  - task:
      type: text-classification
      id: plain
    dataset:
      type: akaruineko/ftanch
      name: FTANch
      split: test
    metrics:
    - type: acc
      value: 0.7012
      name: Accuracy
    - type: p
      value: 0.4647
      name: Precision
    - type: r
      value: 0.8784
      name: Recall
    - type: f1
      value: 0.6078
      name: F1-Score
  - task:
      type: text-classification
      id: mutated
    dataset:
      type: akaruineko/ftanch
      name: FTANch
      split: test
    metrics:
    - type: acc
      value: 0.9134
      name: Accuracy
    - type: p
      value: 0.9152
      name: Precision
    - type: r
      value: 0.9778
      name: Recall
    - type: f1
      value: 0.9455
      name: F1-Score 
---

# ftan-2.5

**ftan-2.5** is a fine-tuned [DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased) sequence classification model for detecting offensive text.

The model predicts one of two labels:

* `clean` — non-offensive text
* `offensive` — offensive text

## Training

ftan-2.5 is the continuation of the [`akaruineko/ftan-2.0`](https://huggingface.co/akaruineko/ftan-2.0) project.

The new version was trained on the [`akaruineko/fantastic-offensive`](https://huggingface.co/datasets/akaruineko/fantastic-offensive) dataset, containing approximately **2.46 million text samples** with a near-balanced distribution between the `clean` and `offensive` classes.

## Usage

```python
from transformers import pipeline

classifier = pipeline(
    "text-classification",
    model="akaruineko/ftan-2.5"
)

result = classifier("you are stupid")
print(result)
```

Example:

```text
[{'label': 'offensive', 'score': 0.997348427772522}]
```

## Intended Use

ftan-2.5 can be used for:

* content moderation
* filtering offensive messages
* dataset preprocessing
* moderation pipelines

## Limitations

This model should **not** be treated as a perfect moderation system.

Offensiveness can depend heavily on context, intent, quotation, sarcasm, reclaimed language, and the surrounding conversation. The model may therefore produce incorrect predictions for ambiguous or context-dependent text.

For example, a sentence discussing an offensive word may still receive a non-trivial offensive score even when the sentence itself is not an insult.

The model also operates on individual text inputs and does not have access to conversation history unless it is explicitly provided as input.

## License

MIT

## Author

Created by **akaruineko**.

This model is the **2.5** continuation of the `ftan` model series.