© 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. 22-GEMMASCOPE-RES-131K
    7. 16342
    Prev
    Next
    INDEX
    Explanations

    movement upward and downward

    np_acts-logits-general · gemini-2.5-flash-lite

    the presence of spatial direction or location terms, like “up,” “downward,” “from,” and “into.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    delning
    -0.98
     herum
    -0.92
    벡
    -0.90
    outFile
    -0.85
     veckan
    -0.85
    ā
    -0.83
    lidad
    -0.82
     Schluss
    -0.82
    cartItems
    -0.82
     such
    -0.80
    POSITIVE LOGITS
     вверх
    1.13
     вниз
    0.96
    marginTop
    0.92
     observaciones
    0.91
    remera
    0.90
    ynx
    0.90
     disait
    0.88
    unii
    0.85
     elevation
    0.85
     élev
    0.84
    Activations Density 0.079%

    No Known Activations