Downloads · 30 days
4
8% of all-time downloads
hansen-grooming-lora-gemma4-e2b
erikaecl/hansen-grooming-lora-gemma4-e2b
hansen-grooming-lora-gemma4-e2b is a text classification model from erikaecl. Use it when you need a label for a piece of text. It is set up for peft. The card lists the license as apache-2.0.
LoRA adapter fine-tuned on top of google/gemma-4-E2B for binary classification of online grooming conversations.
All-time downloads
52
Public
Repo size
55.1 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.json32.2 MB · 58%
At a glance
- Task
- Text Classification
- Library
- peft
- License
- apache-2.0
- Access
- Public
- Created
- May 30, 2026
- Updated
- May 30, 2026
- SHA
- 69f5e759
Try a prompt
Base models
Model card
From the Hugging Face model README
Hansen Grooming LoRA Adapter — Gemma 4 E2B + LoRA (4-bit)
LoRA adapter fine-tuned on top of google/gemma-4-E2B for binary classification of online grooming conversations.
Training Details
- Base model: google/gemma-4-E2B
- Method: QLoRA (4-bit NF4, rank=16, alpha=32)
- Task: Binary sequence classification (Safe vs Grooming)
- Dataset: PAN12 + AOL + NPS Chat + synthetic negatives (anonymized)
Usage
Gemma 4 has no native AutoModelForSequenceClassification. Load with lora_loader.py:
from lora_loader import load_lora_classifier
from grooming_lora_config import build_input
model, tokenizer, device = load_lora_classifier("erikaecl/hansen-grooming-lora-gemma4-e2b", device="cuda")
# model = Gemma4SequenceClassifier (LoRA backbone + classifier.bin head)
Adapter files: PEFT LoRA weights + classifier.bin + classifier_config.json.
- Task
- Text Classification
- Library
- peft
- License
- apache-2.0
- Created
- May 30, 2026
- Updated
- May 30, 2026