© 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. 34-GEMMASCOPE-RES-131K
    7. 48696
    Prev
    Next
    INDEX
    Explanations

    options and settings

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

    The neuron strongly activates on product‐advertising or specification text—i.e. descriptive marketing copy and feature/spec lists (e.g. “comfortable,” “portable,” “features,” wattage settings, dimensions, etc.).

    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
    غة
    -0.84
     asked
    -0.83
     イベント
    -0.82
    нению
    -0.80
    douard
    -0.79
    uſ
    -0.76
    wörter
    -0.75
     gối
    -0.74
     給
    -0.74
    🌅
    -0.73
    POSITIVE LOGITS
     modes
    1.36
     settings
    1.29
    modes
    1.16
    Modes
    1.14
     levels
    1.14
     Modes
    1.14
     choice
    1.13
    choose
    1.11
    Settings
    1.10
    Setting
    1.07
    Activations Density 0.014%

    No Known Activations