© 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. 77700
    Prev
    Next
    INDEX
    Explanations

    more than just

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

    This neuron activates on numeric tokens, especially multi-digit numbers or large numeric values.

    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
    ולים
    -0.96
     largely
    -0.95
    ificacion
    -0.94
     приняли
    -0.94
    ɵɵelementStart
    -0.94
     rather
    -0.94
    such
    -0.93
    婠
    -0.92
    cribir
    -0.91
    ÉD
    -0.91
    POSITIVE LOGITS
     merely
    2.05
     просто
    1.99
     mere
    1.98
     just
    1.84
     только
    1.79
    แค่
    1.75
    只是
    1.73
     مجرد
    1.70
     tylko
    1.68
     simply
    1.60
    Activations Density 0.039%

    No Known Activations