© 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. 27-TRANSCODER-HP
    7. 8672
    Prev
    Next
    INDEX
    Explanations

    back

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     suppressed
    -0.28
     brick
    -0.26
    uits
    -0.25
    lek
    -0.25
    undi
    -0.25
    rios
    -0.24
    ä¸įåĬ¨
    -0.24
     share
    -0.24
    atic
    -0.24
    keiten
    -0.23
    POSITIVE LOGITS
    使人
    0.30
    bud
    0.26
     gover
    0.25
    ÙĬئة
    0.24
     EOF
    0.24
    acious
    0.24
    cura
    0.23
    kur
    0.23
    vere
    0.23
     enact
    0.23
    Activations Density 0.014%

    No Known Activations