© 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. 214214
    Prev
    Next
    INDEX
    Explanations

    offense and defense states

    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
     shooter
    0.50
     handling
    0.42
    )|^{
    0.42
    Handling
    0.41
    handling
    0.40
     iht
    0.40
     HANDLING
    0.40
     graphical
    0.39
     drafting
    0.39
    纷
    0.39
    POSITIVE LOGITS
     improvement
    0.41
    improvement
    0.41
    担忧
    0.39
    anco
    0.38
    Funcion
    0.38
     исто
    0.37
    units
    0.37
     amici
    0.37
    襯
    0.37
     특징
    0.36
    Activations Density 0.006%

    No Known Activations