© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1276
    Prev
    Next
    INDEX
    Explanations

    NY Penal

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

    This neuron fires most strongly on numeric expressions and other quantitative tokens—dates, counts, durations, and amounts—in the text.

    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
    idase
    -1.27
    hline
    -1.20
    本日は
    -1.20
    TestingModule
    -1.19
     esports
    -1.18
    ‚¬
    -1.18
    @[+][
    -1.18
    /**
    -1.15
    Mrs
    -1.13
    ։
    -1.13
    POSITIVE LOGITS
     жы
    1.09
    Innen
    1.02
     compost
    0.99
     people
    0.98
    őle
    0.96
     folks
    0.94
     územ
    0.92
     Oberlin
    0.91
     Bellingham
    0.91
     our
    0.89
    Activations Density 0.181%

    No Known Activations