© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 32883
    Prev
    Next
    INDEX
    Explanations

    public methods and range definitions

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

    The neuron is triggered by tokens that look like numbers or code‐style identifiers (e.g. numeric literals or all‐caps/single‐letter variable names).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    
    -1.10
    ΄
    -1.09
     chè
    -1.06
    ཥ
    -1.05
    opies
    -1.01
    閪
    -1.00
    anity
    -0.98
     profiter
    -0.97
    ckeye
    -0.96
    lity
    -0.96
    POSITIVE LOGITS
     cannot
    1.06
    相关
    1.02
     and
    1.02
    是什么
    1.02
     tells
    0.98
    nö
    0.98
     presumably
    0.94
     μαύ
    0.92
     probably
    0.91
     gets
    0.91
    Activations Density 0.028%

    No Known Activations