© 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. 22-TRANSCODER-HP
    7. 97959
    Prev
    Next
    INDEX
    Explanations

    chief, head

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uv
    -0.28
    vertis
    -0.27
    åĿļå®ŀçļĦ
    -0.27
    éł
    -0.27
    \/\/
    -0.25
    çļĦèī¯å¥½
    -0.25
    èī¯å¥½
    -0.24
    æ¯Ķäºļ
    -0.24
    åĻĶ
    -0.23
    erti
    -0.23
    POSITIVE LOGITS
    among
    0.29
     among
    0.28
     lul
    0.28
     /************************************************************************
    0.27
     Lamar
    0.26
    ikan
    0.25
     blew
    0.25
    左边
    0.25
    alue
    0.24
     nhân
    0.24
    Activations Density 0.096%

    No Known Activations