© 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. 22-GEMMASCOPE-RES-131K
    7. 15657
    Prev
    Next
    INDEX
    Explanations

    issues and conditions

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

    The neuron strongly fires on content‐heavy, technical or domain‐specific nouns (e.g. “intolerances,” “psychopathology,” “condition,” “vector,” “problems,” “NPA,” “ABS”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    terms denoting problems, defects, or undesirable/pathological conditions across technical or medical contexts.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    words related to problems, issues, or negative conditions (often in technical, medical, or structured contexts).

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
    vala
    -1.02
    penis
    -1.02
    inergic
    -1.02
    ådet
    -1.00
     ORIENT
    -0.97
     ´
    -0.96
    шается
    -0.93
    
    -0.93
     sparkles
    -0.92
    до
    -0.91
    POSITIVE LOGITS
     caused
    1.28
     prevention
    1.10
     ellos
    1.09
    Causes
    1.08
     causes
    1.05
     case
    0.99
     severity
    0.99
     Causes
    0.95
     analysis
    0.94
    bilità
    0.93
    Activations Density 0.168%

    No Known Activations