© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 90223
    Prev
    Next
    INDEX
    Explanations

    copyright notices, research goals

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

    The neuron strongly activates on tokens composed of uppercase letters or acronyms (e.g. class names, package/import identifiers) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Eisen
    -0.87
     Lula
    -0.82
    ッシュ
    -0.81
     Gujarati
    -0.80
     HOF
    -0.80
     Banglades
    -0.79
    石头
    -0.78
     brazil
    -0.78
    LOY
    -0.77
     worsened
    -0.77
    POSITIVE LOGITS
     XML
    0.97
    InMillis
    0.84
    XML
    0.84
    верь
    0.84
    淖
    0.83
     engan
    0.79
     Bernal
    0.75
    ثر
    0.74
     viera
    0.74
    gite
    0.72
    Activations Density 0.022%

    No Known Activations