© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5250
    Prev
    Next
    INDEX
    Explanations

    Slang or informal terms

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

    This neuron is strongly activated by individual Hebrew letters or tokens composed of Hebrew characters.

    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
    Vol
    0.55
     바
    0.54
    Barry
    0.52
    바
    0.51
     °
    0.50
    ,//
    0.50
     Ната
    0.49
     Voluntary
    0.49
     monetize
    0.49
     Zuh
    0.49
    POSITIVE LOGITS
    awab
    0.65
     winemaker
    0.63
     tratamento
    0.60
     tratamiento
    0.59
    /*
    0.55
    allar
    0.55
    (/*
    0.54
    丕
    0.54
    itting
    0.53
     /*
    0.52
    Activations Density 0.016%

    No Known Activations