© 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. 31-GEMMASCOPE-2-RES-16K
    7. 13432
    Prev
    Next
    INDEX
    Explanations

    into sexual

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

    The neuron fires on numeric tokens—especially standalone numbers or price‐style values.

    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_31_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
    呃
    0.59
    SACTION
    0.58
    ariski
    0.58
    𝘤
    0.58
    硣
    0.57
     року
    0.56
    闓
    0.56
    ECUTIVE
    0.56
    嗚
    0.56
    ్రియ
    0.55
    POSITIVE LOGITS
     Shape
    0.62
     plateau
    0.61
     shape
    0.56
     vai
    0.55
     get
    0.55
     locom
    0.55
    かず
    0.54
     অবস্থানে
    0.53
     observer
    0.52
     облі
    0.52
    Activations Density 0.000%

    No Known Activations