© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 22157
    Prev
    Next
    INDEX
    Explanations

    reporting what someone said

    np_acts-logits-general · gemini-2.5-flash-lite

    saying or reporting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ب
    0.81
    ב
    0.81
    ه
    0.77
    u
    0.71
    י
    0.71
    ي
    0.70
     -
    0.69
     Cartan
    0.68
    ственный
    0.66
    ления
    0.66
    POSITIVE LOGITS
     goodbye
    0.68
    实话
    0.68
    gı
    0.65
    loud
    0.64
    <0x91>
    0.63
    uti
    0.63
     giusta
    0.63
    ٢
    0.62
    名の
    0.62
     louder
    0.61
    Activations Density 0.082%

    No Known Activations