© 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. 40-GEMMASCOPE-2-RES-262K
    7. 220996
    Prev
    Next
    INDEX
    Explanations

    Traumatic Event, harming, major geopolitical

    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_40_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
     zas
    0.41
     ak
    0.37
     oppure
    0.37
     comod
    0.36
    }.
    0.36
     ironic
    0.36
     hard
    0.36
     fuori
    0.36
     proved
    0.35
     vanno
    0.35
    POSITIVE LOGITS
     conséquence
    0.44
    appendText
    0.43
    飧
    0.41
     Timberwolves
    0.40
    ствие
    0.40
     მთ
    0.40
     sportsman
    0.39
     glTexCoord
    0.39
    Neal
    0.39
    듕
    0.39
    Activations Density 0.000%

    No Known Activations