© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 39-GEMMASCOPE-2-TRANSCODER-262K
    7. 3780
    Prev
    Next
    INDEX
    Explanations

    overthinking or overreacting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_39_width_262k_l0_small_affine
    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
    咙
    0.39
    ریشن
    0.38
    ahuan
    0.37
    ipzig
    0.37
    吣
    0.36
    த்தன
    0.36
     magic
    0.36
     தொழ
    0.36
     സെക്ര
    0.36
    elijk
    0.36
    POSITIVE LOGITS
    grav
    0.37
     Calvary
    0.36
    sini
    0.32
    े
    0.32
     Shepherd
    0.31
    complement
    0.31
     впечат
    0.31
     ट्रॉ
    0.31
    newValue
    0.31
    గిన
    0.31
    Activations Density 0.016%

    No Known Activations