© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13303
    Prev
    Next
    INDEX
    Explanations

    grace and place

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

    The neuron strongly activates on the tokens “place” or “places,” i.e. mentions of that word.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    toare
    0.67
    ത്തിലാണ്
    0.66
    ampu
    0.62
     Генера
    0.61
    ayeva
    0.61
    gebiet
    0.61
     område
    0.60
    ᴊ
    0.59
    ﻕ
    0.59
    STOP
    0.57
    POSITIVE LOGITS
     place
    1.22
     Place
    1.17
     PLACE
    1.11
    place
    1.09
    Place
    1.05
     plaats
    0.89
     lugar
    0.88
     जगह
    0.85
     स्थान
    0.83
     جگہ
    0.83
    Activations Density 0.318%

    No Known Activations