© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112976
    Prev
    Next
    INDEX
    Explanations

    wickedness, crooked teeth, Reilly/Rupert names

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly fires on the word “wicked” (in any casing or morphological variant such as “wickedness”) and, to a lesser degree, on similar “-cked” endings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     midway
    -3.06
    翃
    -2.89
    窳
    -2.61
    店的
    -2.53
    珢
    -2.52
    }$.
    -2.50
    霎
    -2.50
    
    -2.47
    廰
    -2.45
    荜
    -2.44
    POSITIVE LOGITS
    .
    2.94
    the
    2.64
     That
    2.55
    s
    2.36
    iscono
    2.25
    7
    2.25
    ной
    2.22
    ness
    2.22
     Following
    2.19
     इसलिए
    2.19
    Activations Density 0.001%

    No Known Activations