© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 13200
    Prev
    Next
    INDEX
    Explanations

    code syntax

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

    The neuron is triggered by class/type names used in object‐creation expressions (e.g. “new ClassName(...)”).

    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
    ých
    -1.08
     kiezen
    -0.97
     having
    -0.97
    سك
    -0.96
    mpotent
    -0.94
     Ufer
    -0.94
    Each
    -0.93
     cartaz
    -0.92
    return
    -0.91
    ringar
    -0.91
    POSITIVE LOGITS
    操作系统
    1.06
     ملک
    0.96
     coordon
    0.95
     död
    0.93
     said
    0.91
    cipient
    0.90
     бъ
    0.89
     cardin
    0.87
     quintessential
    0.87
     한다
    0.86
    Activations Density 0.007%

    No Known Activations