© 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-3-1B-IT
    5. 17-GEMMASCOPE-2-RES-16K
    7. 4654
    Prev
    Next
    INDEX
    Explanations

    undesirable outcomes and problems

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-1b-it/resid_post/layer_17_width_16k_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
    <unused2204>
    0.78
    ographic
    0.75
    <unused1838>
    0.75
    <unused321>
    0.74
    <unused2136>
    0.74
    <unused638>
    0.73
    <unused693>
    0.72
    <unused2197>
    0.72
    <unused473>
    0.72
    <unused1654>
    0.72
    POSITIVE LOGITS
     kayb
    0.78
     peor
    0.78
     schlecht
    0.75
     ruining
    0.74
     불편
    0.74
     inefficiency
    0.74
     congestion
    0.74
     খারাপ
    0.74
     disadvantage
    0.74
    ndrome
    0.74
    Activations Density 0.456%

    No Known Activations