© 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-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 23078
    Prev
    Next
    INDEX
    Explanations

    establish safe words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    製品
    0.46
     CORPOR
    0.43
     기업
    0.42
     liczba
    0.41
    輻
    0.41
     устанавли
    0.41
    nios
    0.41
     empresa
    0.40
    ၼ
    0.40
    cios
    0.40
    POSITIVE LOGITS
     hilly
    0.46
    BottomOf
    0.46
     competition
    0.45
     showdown
    0.45
     battle
    0.44
     kinetic
    0.43
     hill
    0.42
     scent
    0.42
     misconception
    0.42
    avorites
    0.41
    Activations Density 0.004%

    No Known Activations