© 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. 70691
    Prev
    Next
    INDEX
    Explanations

    gone by or forever

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

    The neuron reliably fires on the word “gone,” picking up on that past‐participle form indicating something has disappeared or is no longer present.

    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
    ஂ
    -2.94
    昞
    -2.30
    0
    -2.27
    ﮨ
    -2.23
    凩
    -2.19
    you
    -2.17
    にほんブログ村
    -2.14
    鷓
    -2.13
    >
    -2.09
    垕
    -2.08
    POSITIVE LOGITS
    ↵
    2.75
    "
    2.33
    י
    2.22
    ，「
    2.19
     Our
    2.17
    ’，
    2.17
    ’
    2.06
    $'
    1.99
     sizzling
    1.95
    కి
    1.95
    Activations Density 0.003%

    No Known Activations