© 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. 65144
    Prev
    Next
    INDEX
    Explanations

    shift, rotate, move, delay, position

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

    The neuron fires on terms that denote shifting, rotating, wrapping, or otherwise moving or translating elements.

    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
    became
    -1.02
    旸
    -0.98
    began
    -0.96
    dislike
    -0.96
     formules
    -0.95
    innovative
    -0.94
    ेट
    -0.94
     explicou
    -0.94
     thẻ
    -0.94
     étoit
    -0.94
    POSITIVE LOGITS
     from
    1.32
     to
    1.26
     chrétienne
    1.08
    操作
    1.06
     ilmiah
    1.04
     praktis
    1.02
     posiciones
    1.00
    capulco
    0.99
    offset
    0.96
     ourselves
    0.96
    Activations Density 0.037%

    No Known Activations