© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2032
    Prev
    Next
    INDEX
    Explanations

    check-in

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

    The neuron fires on the first element of hyphenated or paired multi‐word compounds (e.g. “stand-alone,” “runner-up,” “go-shop,” “feel-good,” “problem-solve,” “road-trip,” etc.).

    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_31_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
    $
    0.55
    Suppress
    0.53
    ールの
    0.52
     Chains
    0.52
    truncate
    0.52
     Conditioner
    0.52
     Convenience
    0.52
     Weird
    0.52
     Mutant
    0.51
    зывать
    0.50
    POSITIVE LOGITS
     expedited
    0.69
     inqui
    0.68
    -
    0.68
     באתר
    0.67
     astăzi
    0.63
     foray
    0.63
    क्की
    0.60
     onboarding
    0.60
     korero
    0.60
    ‒
    0.60
    Activations Density 0.436%

    No Known Activations