© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 3885
    Prev
    Next
    INDEX
    Explanations

    at least one

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

    The neuron strongly activates on numeric tokens—especially floating‐point or large decimal numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     стекло
    0.82
     것입니다
    0.81
    textView
    0.77
    們
    0.77
    ेल
    0.77
    其实
    0.77
     proud
    0.76
     形
    0.76
     其实
    0.73
     Той
    0.73
    POSITIVE LOGITS
    ের
    1.29
    পক্ষে
    1.25
    s
    1.12
    ع
    1.11
    ો
    1.10
    на
    1.09
     Didn
    1.06
    ség
    1.05
    Didn
    1.05
    ों
    1.04
    Activations Density 0.024%

    No Known Activations