© 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. 22-GEMMASCOPE-RES-131K
    7. 34234
    Prev
    Next
    INDEX
    Explanations

    also siblings

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

    The neuron activates on numerical tokens—years, ages, counts, or any digits 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_22/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
     bunun
    -1.16
    𝚒
    -1.05
     ketua
    -1.02
    madeus
    -1.01
    imetsu
    -0.99
     Fußballspieler
    -0.98
     அத
    -0.97
     terbaru
    -0.94
     including
    -0.94
     for
    -0.93
    POSITIVE LOGITS
     ebenfalls
    1.58
     szint
    1.38
     whom
    1.32
     is
    1.31
     also
    1.28
     was
    1.26
     тоже
    1.24
    こちらも
    1.19
     også
    1.18
     attente
    1.09
    Activations Density 0.027%

    No Known Activations