© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13529
    Prev
    Next
    INDEX
    Explanations

    Instructions/Questions

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

    This neuron fires strongly on the very first meaningful token after each new segment starts, effectively marking segment or paragraph beginnings.

    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
     Muh
    0.75
    யூ
    0.72
    }}}=
    0.71
     ecuaciones
    0.68
    EE
    0.67
    }}}
    0.67
    SOCK
    0.66
     phenotypes
    0.66
     llamados
    0.66
     спустя
    0.66
    POSITIVE LOGITS
    sx
    0.81
     mauvaise
    0.80
     énergie
    0.77
    ્વ
    0.73
     excellente
    0.73
     qui
    0.73
     noix
    0.72
     bonne
    0.71
    黾
    0.71
    sz
    0.71
    Activations Density 0.001%

    No Known Activations