© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27794
    Prev
    Next
    INDEX
    Explanations

    dose | irritation | format

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

    methods or functions that operate on data, particularly focusing on string manipulation or formatting operations.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on identifiers used as object or class members—i.e. method and property names accessed via “.” or “->.”

    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
     следу
    -1.08
    ないけど
    -0.98
     две
    -0.98
     jūs
    -0.97
    ینه
    -0.97
     chrétiens
    -0.95
     following
    -0.93
     способы
    -0.92
    יסט
    -0.92
    чан
    -0.91
    POSITIVE LOGITS
     Sometimes
    1.04
    Якщо
    0.97
    Sometimes
    0.97
     oftentimes
    0.96
    رياض
    0.94
    🎮
    0.93
     заканчи
    0.91
     undeniably
    0.91
    ScopeManager
    0.87
     Ultimately
    0.87
    Activations Density 0.000%

    No Known Activations