© 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. 34-GEMMASCOPE-RES-131K
    7. 106608
    Prev
    Next
    INDEX
    Explanations

    Seo Optimization, Rapid changes, Non-volatile

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

    This neuron is responding to where you are in a large document (e.g. at the ends of long code or comment blocks) – in other words, it encodes a token’s overall position in the text.

    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
    queryInterface
    -0.88
    ướp
    -0.85
    voort
    -0.80
    ряду
    -0.79
     département
    -0.78
    Since
    -0.77
     Northeastern
    -0.75
     Aside
    -0.75
    )：
    -0.74
    vende
    -0.73
    POSITIVE LOGITS
    CUP
    1.13
     semplici
    0.89
     acoper
    0.86
     empez
    0.84
    hike
    0.81
     CUP
    0.81
     новый
    0.81
     التعليم
    0.80
     bunt
    0.79
     rotación
    0.78
    Activations Density 0.000%

    No Known Activations