© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111777
    Prev
    Next
    INDEX
    Explanations

    code snippets, examples, or technical terms

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

    Looking at the activations, the neuron strongly activates on: - Chinese characters (对, 对话框) - Programming code tokens like "convertView", "RelativeLayout", "for", "pairs" - Context where variables, parameters, or technical terms are being

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ist
    -2.69
     but
    -2.22
    呱
    -2.20
    ?”
    -2.16
     ”
    -2.11
    izing
    -2.09
    ле
    -2.03
     constaté
    -1.95
    )$.
    -1.95
     используется
    -1.92
    POSITIVE LOGITS
    盌
    2.80
    嚳
    2.28
    欵
    2.27
     kinderen
    2.23
    utrients
    2.23
    celotti
    2.17
    tedì
    2.16
    嘁
    2.13
     allerlei
    2.09
     wszyscy
    2.08
    Activations Density 0.000%

    No Known Activations