© 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. 34-GEMMASCOPE-RES-131K
    7. 57154
    Prev
    Next
    INDEX
    Explanations

    moving on

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

    This neuron detects the token “on” when it’s used as part of a common phrasal verb (e.g. “move on,” “head on”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     on
    -1.20
    を行い
    -1.09
    時間は
    -1.06
    What
    -1.05
    From
    -1.05
    to
    -1.05
    يقة
    -1.02
    Since
    -1.02
    How
    -0.99
    Here
    -0.98
    POSITIVE LOGITS
     behalf
    1.18
     occasion
    1.14
     into
    1.07
    nion
    1.04
    同伴
    1.03
     intermin
    1.02
     to
    1.00
     reprezent
    0.96
     магазина
    0.95
     infuriating
    0.94
    Activations Density 0.023%

    No Known Activations