© 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. 2520
    Prev
    Next
    INDEX
    Explanations

    sign/signature followed by of/de

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

    This neuron detects “sign up”–style call-to-action phrases prompting users to register or subscribe.

    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
    侢
    0.67
    侎
    0.62
    烣
    0.59
     वें
    0.58
    pas
    0.57
    და
    0.55
     essen
    0.55
     ProductCategory
    0.54
    搂
    0.54
    목
    0.54
    POSITIVE LOGITS
     signing
    2.05
     Signing
    2.00
     Sign
    1.96
    Signing
    1.94
     sign
    1.91
    Sign
    1.91
     signed
    1.88
    簽
    1.88
    Signed
    1.86
    sign
    1.84
    Activations Density 0.197%

    No Known Activations