Downloads · 30 days
9
36% of all-time downloads
hw1-hc3-detector
HongjiP/hw1-hc3-detector
hw1-hc3-detector is a text classification model from HongjiP. Use it when you need a label for a piece of text. The card lists the license as apache-2.0.
Fine-tuned sentence-transformers/all-MiniLM-L6-v2 for binary classification of HC3 answers: 0 = human, 1 = ChatGPT.
All-time downloads
25
Public
Parameters
22.7M
90.9 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors90.9 MB · 99%
At a glance
- Task
- Text Classification
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Sep 19, 2026
- Updated
- Sep 19, 2026
- SHA
- 5cdcc1d7
Try a prompt
Base models
Model card
From the Hugging Face model README
HW1 — HC3 AI-Generated Text Detector
Fine-tuned sentence-transformers/all-MiniLM-L6-v2 for binary classification of
HC3 answers: 0 = human, 1 = ChatGPT.
Results (HC3 test split, 80/10/10 question-level split, seed 42)
| Model | Test accuracy |
|---|---|
| Frozen MiniLM embeddings + logistic regression (baseline) | 0.8449 |
| Fine-tuned MiniLM (5 epochs, AdamW, lr 2e-5, max_len 256) | 0.9919 |
Training setup
- Base model:
sentence-transformers/all-MiniLM-L6-v2 - Optimizer: AdamW, lr 2e-5, batch size 32, 5 epochs, max sequence length 256
- Data:
Hello-SimpleAI/HC3(English), revision4d0ff18143b5a7e1b1e79beb540c04549d1e59d3, one human and one ChatGPT answer per question, split 80/10/10 by question so answers to the same question never cross splits.
Limitations
HC3 is a historical benchmark: the ChatGPT answers come from one model in late 2022 / early 2023 and the human answers from web forums, so much of the separable signal is style and formatting specific to those sources. This model is not a reliable detector for current student work or for text from newer or differently-prompted models.
- Task
- Text Classification
- Type
- bert
- License
- apache-2.0
- Created
- Sep 19, 2026
- Updated
- Sep 19, 2026