© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 84293
    Prev
    Next
    INDEX
    Explanations

    efficient, effective, secure, safe

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

    The neuron activates on evaluative adjectives that describe desirable product or system qualities (e.g. safe, efficient, secure, reliable, effective).

    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
     but
    -1.33
    ácii
    -1.09
    ឃ
    -0.99
     مكتبه
    -0.97
     besluit
    -0.95
     zil
    -0.94
    游戏中
    -0.90
    其他
    -0.88
    afari
    -0.87
    ランダ
    -0.87
    POSITIVE LOGITS
    igable
    0.90
     carburant
    0.88
     conster
    0.88
     perpé
    0.85
    礒
    0.85
     triom
    0.85
     н
    0.85
     fundo
    0.85
     alternatives
    0.85
     balans
    0.85
    Activations Density 0.054%

    No Known Activations