© 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. 131691
    Prev
    Next
    INDEX
    Explanations

    cost optimization

    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
    entry
    0.45
     unusable
    0.44
    cash
    0.41
     ঢাকার
    0.41
    ovation
    0.38
    <0x9A>
    0.38
    Value
    0.38
    ernet
    0.38
     cash
    0.37
    entity
    0.37
    POSITIVE LOGITS
     oder
    0.41
     dominates
    0.41
    หรือ
    0.40
     mediators
    0.40
     precludes
    0.39
     McKenna
    0.39
    藠
    0.38
    MCS
    0.38
    odar
    0.38
    的路
    0.38
    Activations Density 0.000%

    No Known Activations