© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 45247
    Prev
    Next
    INDEX
    Explanations

    jumper, EPC, Kotlin

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

    The neuron is essentially a “jumper” detector—it activates strongly on occurrences of the word “jumper.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    </sup>
    -2.70
    }
    -2.42
    秾
    -2.41
    </h4>
    -2.28
    絝
    -2.22
    _
    -2.14
    Hmmm
    -2.11
    of
    -2.08
    ㍉
    -2.05
     которы
    -2.03
    POSITIVE LOGITS
     These
    2.31
     or
    2.27
    —¡
    2.08
     striving
    2.02
     for
    2.00
     of
    2.00
     we
    1.94
     escarp
    1.93
     since
    1.91
    aphne
    1.91
    Activations Density 0.003%

    No Known Activations