Note: This model is still experimental, do not use it in production!
BananaMind-Content-Safety-Mini-1.5 is a small text-only content-safety classifier based on LiquidAI/LFM2.5-350M-Base.
It classifies an input prompt as either Safe or Unsafe and, for unsafe content, returns violated safety categories.
This model is intended for lightweight prompt moderation, safety filtering, and local safety classification.
Model Details
Field
Value
Base model
LiquidAI/LFM2.5-350M-Base
Training method
LoRA fine-tuning, then merged
Trainable LoRA parameters
11,993,088
Total parameters during LoRA training
366,477,056
LoRA rank
32
LoRA alpha
64
Training samples
40,000
Safe samples
15,000
Unsafe samples
25,000
Max sequence length
2048
Dataset
nvidia/Nemotron-3.5-Content-Safety-Dataset
Input type
Text only
Image data
Not used
Response-label data
Not used
Reasoning traces
Not used
Prompt format:
text
1You are a strict content-safety classifier.
23Classify the input as Safe or Unsafe.
45Return exactly one of these formats:
67Safe
89or:
1011Unsafe
12Violated Categories: <categories>
13Input: {text_to_classify}
Target format for safe content:
Safe
Target format for unsafe content:
text
1Unsafe
2Violated Categories: <categories>
Evaluation
Evaluation Results
Evaluated on the text-only test split of nvidia/Nemotron-3.5-Content-Safety-Dataset.
Metric
Value
Rows
3,340
Accuracy
92.01%
Overall error rate
7.99%
Invalid outputs
28
Unsafe precision
93.91%
Unsafe recall
97.38%
Unsafe F1
95.61%
Confusion matrix:
Predicted Unsafe
Predicted Safe
Actually Unsafe
2,605
70
Actually Safe
169
468
The model is optimized for high unsafe recall. It catches most unsafe prompts, but it can over-classify some safe prompts as unsafe.
Comparison
Model
Overall Error Rate
BananaMind V2.5 Content Safety E2B
5.12%
BananaMind-Content-Safety-Mini-1.5
7.99%
ShieldGemma 2B
35.30%
ShieldGemma 9B
40.96%
ShieldGemma 27B 8-bit
52.46%
Usage
python
1import torch
2from transformers import AutoModelForCausalLM, AutoTokenizer
34model_id ="BananaMind/BananaMind-Content-Safety-Mini-1.5"56tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True)7model = AutoModelForCausalLM.from_pretrained(8 model_id,9 dtype=torch.bfloat16 if torch.cuda.is_available()and torch.cuda.is_bf16_supported()else torch.float16,10 device_map="auto",11)1213text_to_classify ="How can I bypass a website login?"1415prompt =f"""You are a strict content-safety classifier.
1617Classify the input as Safe or Unsafe.
1819Return exactly one of these formats:
2021Safe
2223or:
2425Unsafe
26Violated Categories: <categories>
27Input: {text_to_classify}28"""2930inputs = tokenizer(prompt, return_tensors="pt").to(model.device)3132with torch.no_grad():33 output = model.generate(34**inputs,35 max_new_tokens=64,36 do_sample=False,37 pad_token_id=tokenizer.eos_token_id,38 eos_token_id=tokenizer.eos_token_id,39)4041prediction = tokenizer.decode(42 output[0][inputs["input_ids"].shape[-1]:],43 skip_special_tokens=True,44).strip()4546print(prediction)
Example unsafe output:
text
1Unsafe
2Violated Categories: Cyber
Safe example:
text_to_classify = "How do I bake banana bread?"
Expected output:
Safe
Training Data
This model was trained on text-only rows from nvidia/Nemotron-3.5-Content-Safety-Dataset.
License
This model is based on LiquidAI/LFM2.5-350M-Base, which uses the LFM Open License v1.0.
The training dataset is nvidia/Nemotron-3.5-Content-Safety-Dataset.
See THIRD_PARTY_LICENSES.md for more info.
Training Time
We trained this model on a 5070 Ti in about 20 minutes
