© 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. 15327
    Prev
    Next
    INDEX
    Explanations

    2. Background of

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

    The neuron activates on directional movement terms in dance or step-by-step instructions (e.g. “left,” “right,” “step,” “side,” “heel,” “back,” “cross”).

    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
     jPanel
    -1.02
     },
    -0.99
    ं
    -0.98
     JOptionPane
    -0.98
    ๕
    -0.97
     would
    -0.97
    thenia
    -0.96
    هاب
    -0.96
    cile
    -0.95
     ModelAndView
    -0.95
    POSITIVE LOGITS
     хоте
    1.12
    薄い
    1.09
    横に
    0.97
    举例
    0.96
    Climat
    0.95
     exposé
    0.94
    色んな
    0.94
    摁
    0.94
     synthesizer
    0.90
    落ちる
    0.90
    Activations Density 0.062%

    No Known Activations