© 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. 68125
    Prev
    Next
    INDEX
    Explanations

    Something followed by phrases

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

    The neuron responds to occurrences of the filler word “something” (and closely related variants like “else”) in the text.

    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
     menant
    -2.02
    
    -2.00
    ementara
    -1.82
     constater
    -1.82
     pratiquement
    -1.79
     horloge
    -1.79
     plafon
    -1.77
     progressivement
    -1.73
     pengar
    -1.73
    蹀
    -1.73
    POSITIVE LOGITS
     at
    2.61
     something
    2.55
     about
    2.34
    </h3>
    2.31
     that
    2.22
     if
    2.06
     some
    2.00
     as
    1.98
     which
    1.98
     or
    1.97
    Activations Density 0.128%

    No Known Activations