© 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. 10-GEMMASCOPE-RES-131K
    7. 59130
    Prev
    Next
    INDEX
    Explanations

    Eliza, NES, Paj, exer, Eminem, Acorn

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

    The neuron fires on capitalized or all-caps tokens that are specific names, acronyms, or identifiers (e.g. “Calibri,” “AlterField,” “NES,” “Eliza”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    >
    -2.77
    er
    -2.45
    0
    -2.44
     têm
    -2.34
    鶫
    -2.28
     também
    -2.20
    糴
    -2.19
     outras
    -2.13
    '
    -2.13
    self
    -2.11
    POSITIVE LOGITS
     untuk
    2.75
    創意
    2.53
    茀
    2.52
    }$.
    2.47
    杙
    2.41
     ждет
    2.36
     الفلسطيني
    2.33
    ꯸
    2.25
    ⮜
    2.25
    嗚
    2.23
    Activations Density 0.011%

    No Known Activations