© 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. 25965
    Prev
    Next
    INDEX
    Explanations

    mathematical and code structures

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

    The neuron fires strongly on isolated technical keywords—especially names of data structures or formal elements (e.g. “indices,” “array,” “as,” “group,” “)”).

    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
     jenner
    -0.85
     nationaux
    -0.83
    就把
    -0.81
     CAI
    -0.81
    ervices
    -0.80
    ใ
    -0.80
    ://
    -0.79
     verschil
    -0.79
    快要
    -0.78
    üten
    -0.78
    POSITIVE LOGITS
    udara
    0.84
    komp
    0.84
     toutefois
    0.84
    enerys
    0.83
    vaart
    0.83
     declara
    0.81
    ンテナ
    0.81
    被告
    0.78
    ekk
    0.78
    dü
    0.77
    Activations Density 0.156%

    No Known Activations