© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71134
    Prev
    Next
    INDEX
    Explanations

    you probably

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

    The neuron strongly activates on direct second‐person address words—especially “you,” “your,” and conditional starters like “if you” or contrastive “but if you.”

    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
     successivo
    -1.09
     rafael
    -1.05
    тном
    -1.00
     cualquier
    -0.96
    oughby
    -0.96
    uerung
    -0.95
    lahoma
    -0.95
    uté
    -0.94
    titian
    -0.93
     only
    -0.93
    POSITIVE LOGITS
     your
    2.33
     you
    2.11
     yourself
    1.59
     kanskje
    1.38
     misschien
    1.38
    probably
    1.36
     already
    1.34
     vielleicht
    1.34
    either
    1.31
     or
    1.29
    Activations Density 0.072%

    No Known Activations