© 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. 45055
    Prev
    Next
    INDEX
    Explanations

    "in the way", names, or specific entries

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

    The neuron activates strongly on the exact phrase “in the way,” marking occurrences of that obstructive expression.

    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
    There
    -1.97
    #
    -1.86
    Still
    -1.84
    Many
    -1.84
    Almost
    -1.81
    ,
    -1.79
    Some
    -1.77
    You
    -1.74
    Two
    -1.73
    One
    -1.71
    POSITIVE LOGITS
    妧
    1.92
    
    1.90
    </strong>
    1.84
    皞
    1.80
    檍
    1.78
    鶫
    1.77
    webElement
    1.68
    StoryboardSegue
    1.67
    brücke
    1.66
     Menurut
    1.62
    Activations Density 0.004%

    No Known Activations