© 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. 111583
    Prev
    Next
    INDEX
    Explanations

    ratings and rankings

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

    The neuron activates on the term “ratings,” especially when it appears as part of numeric audience or review score contexts.

    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
     When
    -3.33
     Despite
    -3.14
     Although
    -3.00
     Putting
    -2.77
     Whether
    -2.69
     Though
    -2.69
     Getting
    -2.69
     or
    -2.67
     när
    -2.66
     However
    -2.64
    POSITIVE LOGITS
    ところは
    2.77
    部分は
    2.64
    </strong>
    2.59
    ほうが
    2.59
    ↵
    2.52
     stück
    2.48
     agujero
    2.47
    部分が
    2.38
     irres
    2.34
     gewi
    2.34
    Activations Density 0.002%

    No Known Activations