© 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. 54477
    Prev
    Next
    INDEX
    Explanations

    predict actions, discriminated against, decent competitors

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

    This neuron strongly activates on modal interrogatives—questions phrased with “can” or “will” about doing something (e.g. “Can we prevent crimes…?”, “Will we use AI to predict…?”).

    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
    🐾
    -1.63
    窰
    -1.48
    😻
    -1.48
    🍭
    -1.48
     hech
    -1.48
     PRICES
    -1.45
    绢
    -1.43
    豋
    -1.43
     toul
    -1.40
    ţia
    -1.39
    POSITIVE LOGITS
     who
    1.90
     people
    1.63
     например
    1.51
     things
    1.48
     aren
    1.45
     людей
    1.45
    agascar
    1.41
     selben
    1.41
     that
    1.41
     множество
    1.40
    Activations Density 0.132%

    No Known Activations