© 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. 117745
    Prev
    Next
    INDEX
    Explanations

    numbers of games in a series

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

    The neuron strongly activates on isolated numeric tokens—especially digits and digit‐dash‐digit sequences—used to denote scores, game numbers, or series counts.

    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
     zdol
    -0.93
     rubs
    -0.85
     impresionantes
    -0.85
     trimestre
    -0.83
    området
    -0.82
    たい
    -0.79
     mingled
    -0.79
     Wob
    -0.78
    нік
    -0.78
    っていない
    -0.76
    POSITIVE LOGITS
     deciding
    1.33
     Game
    1.03
    必
    0.96
     реша
    0.95
     decides
    0.95
     decisive
    0.92
     Necessary
    0.89
     must
    0.86
     必
    0.83
     clin
    0.82
    Activations Density 0.022%

    No Known Activations