© 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-65K
    7. 50340
    Prev
    Next
    INDEX
    Explanations

    causes and effects

    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_65k_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
    ern
    0.41
     करावा
    0.40
    ленной
    0.39
     Perform
    0.39
     සිය
    0.38
     ]];
    0.38
     dissati
    0.38
     аз
    0.37
    erné
    0.37
    ять
    0.37
    POSITIVE LOGITS
    有利于
    0.65
    有助于
    0.65
    体现
    0.61
     khiến
    0.59
    增加了
    0.58
    吸引
    0.56
     باعث
    0.56
    会让
    0.56
    无疑
    0.55
    能
    0.55
    Activations Density 0.012%

    No Known Activations