© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78823
    Prev
    Next
    INDEX
    Explanations

    dark secrets and corruption

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    𝐏
    -0.84
     হয়
    -0.84
    方案
    -0.82
    AY
    -0.79
    𝐅
    -0.78
     clues
    -0.76
    JsonProperty
    -0.76
    ဗ
    -0.76
    ay
    -0.75
    /</
    -0.75
    POSITIVE LOGITS
     dark
    3.30
     darker
    3.27
    dark
    2.75
     DARK
    2.36
    Dark
    2.25
     Dark
    2.20
    DARK
    2.11
     ugly
    2.03
     darkest
    1.97
     darkness
    1.83
    Activations Density 0.032%

    No Known Activations