© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71785
    Prev
    Next
    INDEX
    Explanations

    possibility or potential futures

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

    The neuron fires on plural noun forms—words referring to multiple items (e.g. “Cheerios,” “Wheaties,” “sharks,” “walls,” “thousands”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     nel
    -1.05
    大人
    -0.99
     Februari
    -0.96
     eski
    -0.96
     Januari
    -0.94
     prend
    -0.93
     self
    -0.93
    显示
    -0.91
    мани
    -0.91
    стить
    -0.91
    POSITIVE LOGITS
     probabilmente
    1.33
    可能會
    1.27
     possibly
    1.25
     potentially
    1.23
     conceivably
    1.16
     potrebbero
    1.14
     kanskje
    1.13
    Probably
    1.13
    まさか
    1.13
     möglicherweise
    1.09
    Activations Density 0.025%

    No Known Activations