© 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. 12-TRANSCODER-HP
    7. 22478
    Prev
    Next
    INDEX
    Explanations

    sugar

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÑĪек
    -0.27
    aland
    -0.26
    (typeof
    -0.26
     tumble
    -0.25
     duro
    -0.25
    .getID
    -0.24
     Tough
    -0.24
    ä¸¤åĽ½
    -0.24
    èĻij
    -0.24
    请æĤ¨
    -0.24
    POSITIVE LOGITS
     fet
    0.28
     Concent
    0.27
     valley
    0.27
     concentrated
    0.26
    trial
    0.26
     concentr
    0.25
    ulg
    0.25
     mái
    0.24
    gross
    0.24
    åij¼ãģ°
    0.24
    Activations Density 0.002%

    No Known Activations