© 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. 2568
    Prev
    Next
    INDEX
    Explanations

    avoid generating harmful content

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

    the word “avoid.”

    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
     здоровья
    0.98
     œuvre
    0.97
    傈
    0.95
     полномо
    0.94
     أد
    0.90
     avulla
    0.90
     ischa
    0.89
    쫌
    0.89
     осно
    0.88
    í
    0.88
    POSITIVE LOGITS
     pitfalls
    1.16
    諱
    1.16
    麻煩
    1.06
    نك
    1.04
    congestion
    1.02
     Czę
    1.02
    讳
    1.02
    ের
    1.01
     wasting
    1.00
     blaming
    1.00
    Activations Density 0.011%

    No Known Activations