© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14726
    Prev
    Next
    INDEX
    Explanations

    "nature" or "elements"

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

    This neuron selectively fires on longer, morphologically complex content words (e.g. multi-syllable nouns and adjectives) rather than common short function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    words or phrases enclosed in quotation marks.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    words or short phrases enclosed in quotation marks, especially scare-quoted terms used for emphasis or irony.

    oai_token-act-pair · gpt-5Triggered 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
    Amit
    0.95
    。",
    0.82
     Amit
    0.81
    Deputy
    0.80
    Ginger
    0.80
     Appl
    0.79
    Municipal
    0.79
    Aye
    0.79
    Drone
    0.79
    Fut
    0.79
    POSITIVE LOGITS
     (
    0.76
    ৭০
    0.73
     (?)
    0.73
     astronomical
    0.70
     percentage
    0.69
     metaphorical
    0.69
    ৭
    0.68
    ৯০
    0.68
     pseudo
    0.67
     conglomerate
    0.67
    Activations Density 0.000%

    No Known Activations