© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 84309
    Prev
    Next
    INDEX
    Explanations

    choosing to add options

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

    The neuron strongly responds to promotional calls‐to‐action—imperative marketing phrases (e.g. “Click HERE to…,” “choose your options,” “order each side”) that instruct the reader what to do.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     شروع
    -0.92
     Vs
    -0.85
    ēl
    -0.84
    ownie
    -0.83
    IAction
    -0.82
    wie
    -0.82
     войне
    -0.81
    ());
    
    -0.81
    !");
    
    -0.79
    uis
    -0.78
    POSITIVE LOGITS
     specify
    1.68
     choose
    1.63
     indicate
    1.51
     select
    1.41
     option
    1.34
     add
    1.32
     choosing
    1.31
    オプション
    1.31
     selecting
    1.29
     specifying
    1.28
    Activations Density 0.032%

    No Known Activations