© 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. 48447
    Prev
    Next
    INDEX
    Explanations

    images in file paths

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

    The neuron strongly activates on numeric tokens (digits, version numbers, years, etc.) 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
     BUILD
    -0.73
    ]='\
    -0.71
     posicion
    -0.68
    поле
    -0.66
    kval
    -0.66
     pemas
    -0.65
    に入れて
    -0.65
    շ
    -0.65
    retrato
    -0.64
     Neub
    -0.63
    POSITIVE LOGITS
    ("'"
    0.71
    illants
    0.70
     Blanket
    0.66
     similaire
    0.65
     Glance
    0.63
    émica
    0.63
     Bali
    0.63
     conso
    0.62
    ullan
    0.61
    弹
    0.61
    Activations Density 0.069%

    No Known Activations