© 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. 65207
    Prev
    Next
    INDEX
    Explanations

    example: input, output, array

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

    This neuron detects the example sections of coding questions, especially the numerical input/output and illustrative sample values.

    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
     and
    -1.07
    datos
    -1.02
     that
    -0.93
     bruke
    -0.92
    5
    -0.91
    mListener
    -0.91
     or
    -0.90
    6
    -0.90
    ,
    -0.90
    3
    -0.89
    POSITIVE LOGITS
     Например
    1.23
     abnorm
    1.21
     經典
    1.16
     結果
    1.13
     darbo
    1.12
     šo
    1.10
     exemplary
    1.07
     nachher
    1.07
    Jeśli
    1.05
     doba
    1.05
    Activations Density 0.155%

    No Known Activations