© 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. 26478
    Prev
    Next
    INDEX
    Explanations

    sha followed by digits

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

    This neuron detects occurrences of the word “shading” (and its close variants like “shaded”) in text.

    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
     mistaken
    -2.77
     entende
    -2.56
     ,‬
    -2.52
    팹
    -2.50
     spesa
    -2.44
     compel
    -2.39
     garantia
    -2.28
     apprehensive
    -2.27
     でも
    -2.22
     stimulates
    -2.20
    POSITIVE LOGITS
    縠
    3.06
    ?
    2.80
    *
    2.63
     nogal
    2.50
     krim
    2.45
     nadru
    2.42
     kram
    2.41
    𞤢
    2.38
     biografi
    2.36
     eksklu
    2.36
    Activations Density 0.017%

    No Known Activations