© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4058
    Prev
    Next
    INDEX
    Explanations

    generating explicit adult content

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

    The neuron strongly activates on content-bearing (non-stop) words—i.e. the uncommon nouns and verbs that carry the main semantic load of a sentence.

    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
    ݢ
    0.87
     कमीशन
    0.81
     jLabel
    0.79
    ِي
    0.78
     掛
    0.78
    ر
    0.77
    சொ
    0.76
    شتہ
    0.75
    要知道
    0.73
    ﭑ
    0.73
    POSITIVE LOGITS
    m
    0.73
     toxicology
    0.71
     “
    0.68
    <0xC2>
    0.67
    VB
    0.67
    mic
    0.66
    itivity
    0.66
    вары
    0.66
    BR
    0.66
     CASA
    0.66
    Activations Density 0.000%

    No Known Activations