© 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. 12916
    Prev
    Next
    INDEX
    Explanations

    turn on or off

    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
    <unused2125>
    0.30
    ма
    0.30
    öyle
    0.30
    <unused2121>
    0.30
     депута
    0.30
     मुसलमान
    0.30
    <unused2178>
    0.30
    ública
    0.29
    <unused1096>
    0.29
    <unused370>
    0.29
    POSITIVE LOGITS
     (
    0.46
     vs
    0.43
     versus
    0.43
     (+)
    0.42
     wartość
    0.41
     logics
    0.40
     và
    0.40
     robustness
    0.39
     variability
    0.39
     hoặc
    0.38
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.