© 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. 8416
    Prev
    Next
    INDEX
    Explanations

    dates, REGISTRATION, premium

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

    This neuron detects infinitive verbs following “to” (e.g. “to simplify,” “to wire”), marking instructional steps.

    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
    רה
    0.80
    పోయ
    0.80
     რამდენ
    0.76
     Administración
    0.75
     tenido
    0.74
     windowsill
    0.74
     semia
    0.74
    敺
    0.74
     propria
    0.73
    <unused2121>
    0.73
    POSITIVE LOGITS
     Them
    0.73
     اخرى
    0.73
     Fertility
    0.73
    ure
    0.72
     load
    0.72
    Load
    0.71
    IR
    0.68
     This
    0.67
    Print
    0.66
     Dieser
    0.66
    Activations Density 0.000%

    No Known Activations