© 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. 106778
    Prev
    Next
    INDEX
    Explanations

    javascript code modules

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

    The neuron responds strongly to numeric tokens—i.e. any sequence of digits.

    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
     inconvénients
    -1.02
     Gales
    -0.95
     Peoria
    -0.94
     congreso
    -0.89
    ついで
    -0.88
     Boun
    -0.87
    Sieg
    -0.85
    心跳
    -0.85
     Personas
    -0.84
     Tequila
    -0.84
    POSITIVE LOGITS
    messageInfo
    1.02
    ➩
    0.96
     onOptions
    0.96
     those
    0.95
     even
    0.90
     variedades
    0.90
    MessageState
    0.90
    čný
    0.90
    ستار
    0.89
    woman
    0.88
    Activations Density 0.084%

    No Known Activations