© 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. 1389
    Prev
    Next
    INDEX
    Explanations

    by + verb

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

    The neuron strongly activates on marketing‐style, call-to-action verbs (e.g. “live,” “providing,” “maximizing,” “exploring,” “informing”) used in promotional or corporate context.

    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
     бывает
    0.70
     wou
    0.70
     simonsen
    0.67
     gelişmeler
    0.66
    ש
    0.66
     previsto
    0.63
     previstas
    0.63
     пройдет
    0.63
     бывают
    0.61
     পাকিস্তানী
    0.60
    POSITIVE LOGITS
     virtue
    0.78
    zantine
    0.76
     combining
    0.69
    ll
    0.68
     clicking
    0.67
    ಿಲ್ಲ
    0.67
    ಾಗ
    0.66
    ه
    0.64
    te
    0.63
     placing
    0.62
    Activations Density 0.062%

    No Known Activations