© 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. 24-GEMMASCOPE-2-RES-16K
    7. 2273
    Prev
    Next
    INDEX
    Explanations

    concepts after "of"

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

    The neuron fires strongly on unusually rare or “out‐of‐vocabulary” tokens (e.g. long names, uncommon terms, or embedded numeric identifiers).

    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_24_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
     this
    0.82
     the
    0.79
     thefe
    0.72
     these
    0.70
     theses
    0.69
     এই
    0.67
     a
    0.65
     an
    0.64
     இந்த
    0.63
     acest
    0.60
    POSITIVE LOGITS
     extraordinaire
    0.59
    renown
    0.56
     ambayo
    0.56
    𝓭
    0.56
    𝗿
    0.55
    𝒓
    0.55
     tujuh
    0.54
    circa
    0.54
    𝗹
    0.54
    ine
    0.54
    Activations Density 0.219%

    No Known Activations