© 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. 22-GEMMASCOPE-RES-131K
    7. 72413
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions with powers and variables

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

    The neuron activates on exponentiation expressions (tokens indicating a “**” power operation, e.g. x**n).

    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
    Ḳ
    -0.99
     competir
    -0.91
     różne
    -0.85
     localidades
    -0.85
     noticeable
    -0.85
    さんへ
    -0.84
     and
    -0.84
     what
    -0.82
     dostępne
    -0.82
     mochilas
    -0.82
    POSITIVE LOGITS
     streamlining
    0.94
     IH
    0.92
     to
    0.91
    却
    0.90
     интерьер
    0.90
     adv
    0.87
     formid
    0.85
     siz
    0.84
     samp
    0.84
    SelectAll
    0.83
    Activations Density 0.046%

    No Known Activations