© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15941
    Prev
    Next
    INDEX
    Explanations

    left or right navigation

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

    The neuron strongly fires on adverbial modifiers—especially words like “up” and “fully” that intensify or complete an action.

    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
    נ
    0.86
    ві
    0.86
    في
    0.86
    ن
    0.84
    ною
    0.80
    cec
    0.79
    cea
    0.77
    andı
    0.77
    ק
    0.77
    ري
    0.76
    POSITIVE LOGITS
     thicken
    0.80
     पहुँचा
    0.80
     harass
    0.78
     ardu
    0.77
     brengen
    0.77
     enl
    0.77
     aange
    0.75
     revol
    0.74
     auxili
    0.74
     reminis
    0.74
    Activations Density 0.000%

    No Known Activations