© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 63458
    Prev
    Next
    INDEX
    Explanations

    Non-natural language content

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    opo
    -0.26
    ulous
    -0.26
    è¢ĸ
    -0.24
    cloth
    -0.24
    ddy
    -0.24
    èĤĨ
    -0.23
     Lair
    -0.23
    fox
    -0.23
    <article
    -0.23
    è¨Ģ论
    -0.23
    POSITIVE LOGITS
    gram
    0.27
    ä»Ĺ
    0.27
    å·²ç»ıè¶ħè¿ĩ
    0.27
    ä¹Łå¥½
    0.26
    GRAM
    0.26
     aug
    0.25
     spells
    0.25
     cad
    0.25
     ott
    0.25
     Malk
    0.24
    Activations Density 2.877%

    No Known Activations