© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4422
    Prev
    Next
    INDEX
    Explanations

    animals and objects

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

    This neuron specializes in spotting non-word tokens—especially numeric strings and punctuation sequences (e.g. ellipses, double dots, standalone numbers).

    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
    r
    0.83
     stalwart
    0.74
    '
    0.72
    T
    0.71
     اہلِ
    0.70
     unquestionably
    0.69
     thoughtfully
    0.68
    頗
    0.68
     unanticipated
    0.67
    an
    0.66
    POSITIVE LOGITS
     hidrógeno
    1.05
     bacterias
    1.03
     cytoplas
    1.02
     animales
    0.95
     moléculas
    0.95
     acero
    0.94
     invertebr
    0.93
     existencia
    0.91
     actinides
    0.91
     conoscere
    0.91
    Activations Density 0.031%

    No Known Activations