© 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. 48667
    Prev
    Next
    INDEX
    Explanations

    What is

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

    The neuron consistently lights up on unusually large or rare token IDs (i.e. uncommon/low‐frequency tokens), rather than on any particular semantic content.

    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
    Mockito
    -0.93
     Liberia
    -0.73
    LIB
    -0.73
    ambo
    -0.71
     kitap
    -0.69
    anadas
    -0.69
     Fidelity
    -0.69
    urta
    -0.69
     Laing
    -0.68
    vera
    -0.67
    POSITIVE LOGITS
     Tattoo
    0.71
     Bluffs
    0.65
    Objetivo
    0.64
     euer
    0.63
    气息
    0.62
    endaft
    0.62
    不能
    0.62
    ICP
    0.61
     Atas
    0.61
     Roots
    0.60
    Activations Density 0.046%

    No Known Activations