Implementation:Norrrrrrr lyn WAInjectBench SentenceTransformer Init

Knowledge Sources	WAInjectBench Sentence Transformers
Domains	NLP, Representation_Learning
Last Updated	2026-02-14 16:00 GMT

Overview

Concrete tool for loading the all-MiniLM-L6-v2 sentence embedding model, provided by the sentence-transformers library as used in the WAInjectBench text embedding trainer.

Description

The text embedding training script initializes a SentenceTransformer instance with the model ID "sentence-transformers/all-MiniLM-L6-v2". This model automatically downloads from HuggingFace Hub on first use and produces 384-dimensional embeddings for input text.

Usage

Initialize once before processing multiple JSONL training files. The model is shared across all classifier training runs in a single invocation.

Code Reference

Source Location

• Repository: WAInjectBench
• File: train/embedding-t.py (L58)

Signature

embedder = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")

Import

from sentence_transformers import SentenceTransformer

I/O Contract

Inputs

Name	Type	Required	Description
model_name	str	Yes	HuggingFace model ID (hardcoded: "sentence-transformers/all-MiniLM-L6-v2")

Outputs

Name	Type	Description
embedder	SentenceTransformer	Loaded model instance ready for encode() calls, producing 384-dim vectors

Usage Examples

Initializing the Text Embedder

from sentence_transformers import SentenceTransformer

# Load the model (downloads on first use)
embedder = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")

# Verify output dimensionality
test_emb = embedder.encode(["test sentence"])
print(f"Embedding dim: {test_emb.shape[1]}")  # 384

Related Pages

Implements Principle

• Principle:Norrrrrrr_lyn_WAInjectBench_Text_Embedding_Initialization

Requires Environment

• Environment:Norrrrrrr_lyn_WAInjectBench_Conda_Python_39_CUDA_Environment