© 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. 13403
    Prev
    Next
    INDEX
    Explanations

    To

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

    The neuron fires on lines that begin with “To” or “For,” i.e. the introductory prepositions that mark instruction or guideline steps.

    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
    Ο
    0.60
     смысле
    0.57
     কেবল
    0.57
    這一
    0.55
     것은
    0.54
    stehen
    0.52
    기는
    0.52
    𝙃
    0.52
     Globo
    0.51
     Bologna
    0.50
    POSITIVE LOGITS
     Weitere
    0.63
     សូម
    0.62
    ternut
    0.61
    س
    0.59
     மேலும்
    0.55
     numerals
    0.55
     Autres
    0.55
     referenties
    0.53
     đạp
    0.53
     legible
    0.53
    Activations Density 0.188%

    No Known Activations