© 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-3-4B-IT
    5. 29-GEMMASCOPE-2-RES-262K
    7. 146302
    Prev
    Next
    INDEX
    Explanations

    considers, turn, turn, serve

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_29_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
     உர
    0.76
     idées
    0.72
    麟
    0.70
    かれて
    0.68
     apparaît
    0.68
     parol
    0.67
    िल्
    0.67
    のも
    0.66
     समानता
    0.66
    ）、
    0.65
    POSITIVE LOGITS
     endanger
    0.74
     भीम
    0.70
    ጅ
    0.68
     Callaghan
    0.67
    ⒈
    0.66
    浔
    0.65
     इंद्र
    0.63
    hara
    0.63
     hopeless
    0.63
     progressive
    0.63
    Activations Density 0.002%

    No Known Activations