Downloads · 30 days
23
13% of all-time downloads
ahczhg/ernie-4.5-0.3b-aegis-safety-lora
ernie-4.5-0.3b-aegis-safety-lora is a text classification model from ahczhg. Use it when you need a label for a piece of text. It is set up for transformers. The card lists the license as apache-2.0.
This is a fine-tuned version of Baidu's ERNIE-4.5-0.3B-Base, optimized for content safety applications. ERNIE (Enhanced Representation through kNowledge IntEgration) is particularly strong in understanding context and…
Downloads · 30 days
23
13% of all-time downloads
All-time downloads
174
Public
Parameters
361M
734 MB on disk
Likes
2
Public
Click a slice to open those files.
.safetensors722 MB · 98%
From the Hugging Face model README
This is a fine-tuned version of Baidu's ERNIE-4.5-0.3B-Base, optimized for content safety applications. ERNIE (Enhanced Representation through kNowledge IntEgration) is particularly strong in understanding context and knowledge-enhanced language understanding, making it ideal for nuanced safety detection.
This model was fine-tuned using LoRA (Low-Rank Adaptation) on the NVIDIA Aegis AI Content Safety Dataset 2.0, which contains diverse examples of safe and unsafe content across multiple categories.
pip install transformers torch peft
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
# Load model and tokenizer
model_name = "ahczhg/ernie-4.5-0.3b-aegis-safety-lora"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.float16,
device_map="auto"
)
# Example: Content safety check
prompt = "### Instruction:\nAnalyze this content for safety: 'Your text here'\n\n### Response:\n"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
outputs = model.generate(
**inputs,
max_new_tokens=128,
temperature=0.7,
do_sample=True,
top_p=0.95
)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)
from transformers import pipeline
# Create text generation pipeline
generator = pipeline(
"text-generation",
model="ahczhg/ernie-4.5-0.3b-aegis-safety-lora",
torch_dtype=torch.float16,
device_map="auto"
)
# Generate safety analysis
result = generator(
"### Instruction:\nIs this content safe? 'Hello, how are you?'\n\n### Response:\n",
max_new_tokens=128,
temperature=0.7,
do_sample=True
)
print(result[0]['generated_text'])
The model was evaluated on a held-out test set from the Aegis AI Content Safety Dataset. Key metrics include:
The model was fine-tuned on the NVIDIA Aegis AI Content Safety Dataset 2.0, which includes:
If you use this model in your research or applications, please cite:
@misc{ernie_0.3b_aegis_safety,
author = {ahczhg},
title = {ERNIE-4.5-0.3B Fine-tuned on Aegis AI Content Safety},
year = {2025},
publisher = {HuggingFace},
howpublished = {\url{https://huggingface.co/ahczhg/ernie-4.5-0.3b-aegis-safety-lora}},
note = {Fine-tuned on NVIDIA Aegis AI Content Safety Dataset 2.0}
}
For questions, issues, or feedback, please visit the model repository.