© 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. 31-GEMMASCOPE-2-RES-16K
    7. 7381
    Prev
    Next
    INDEX
    Explanations

    spam/not spam

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

    This neuron detects email-related call-to-action instructions, especially “click the link in the email to confirm/complete a subscription.”

    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
    垫
    0.60
     visas
    0.56
     आक
    0.54
     aspectRatio
    0.53
    žití
    0.53
     Axis
    0.52
     Garage
    0.51
     fabricate
    0.51
    Ŝ
    0.51
    面の
    0.50
    POSITIVE LOGITS
     newsletter
    2.53
     newsletters
    2.38
     Newsletter
    2.33
    newsletter
    2.33
    Newsletter
    2.27
     mailing
    2.05
     Mailing
    1.84
    mailing
    1.66
    subscribe
    1.62
    メール
    1.56
    Activations Density 0.622%

    No Known Activations