© 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. 22-GEMMASCOPE-RES-131K
    7. 128610
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on verbs in their –ed or –ing (participle/gerund) form that mark mechanisms or processes (e.g. “is governed by,” “appears to be concentration‐dependent,” “using,” “performed”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    governed by, created by, depend on

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     eventually
    -0.99
    eventually
    -0.96
    翊
    -0.91
    freude
    -0.90
    ure
    -0.89
     simili
    -0.87
     wist
    -0.84
     Eventually
    -0.83
    technik
    -0.82
    TEMA
    -0.82
    POSITIVE LOGITS
     atacado
    1.09
     аппети
    1.05
     pamamagitan
    1.03
     daarvoor
    1.02
     Commentaires
    0.99
    nother
    0.97
     methods
    0.97
     from
    0.97
    pierw
    0.97
     frasco
    0.95
    Activations Density 0.161%

    No Known Activations