© 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. 33944
    Prev
    Next
    INDEX
    Explanations

    actions and their outcomes

    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
    ).
    0.86
    。
    0.85
    *.
    0.75
    ।
    0.71
    》。
    0.71
     کی۔
    0.70
    ].
    0.70
    ։
    0.70
    ）。
    0.69
    .
    0.69
    POSITIVE LOGITS
     will
    0.73
     would
    0.69
     ayaa
    0.64
     has
    0.59
     could
    0.59
    することは
    0.59
     θα
    0.58
     ก็
    0.57
    就會
    0.56
     may
    0.55
    Activations Density 0.027%

    No Known Activations