© 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. 15520
    Prev
    Next
    INDEX
    Explanations

    short name or description

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

    This neuron consistently lights up on the very first content word of each new segment or section (e.g. the heading or the first word after a break).

    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
    не
    1.11
    ко
    0.82
    ла
    0.81
    де
    0.79
     convergent
    0.78
    রের
    0.76
    ре
    0.74
    ജ
    0.72
    та
    0.71
    на
    0.70
    POSITIVE LOGITS
     Namun
    0.96
     merek
    0.91
     morreu
    0.88
     resultó
    0.88
     rencana
    0.86
     sejak
    0.85
    σουν
    0.85
     работал
    0.84
     انشاء
    0.84
     volvió
    0.83
    Activations Density 0.000%

    No Known Activations