© 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. 65148
    Prev
    Next
    INDEX
    Explanations

    negative outcomes or cancellations

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

    The neuron fires on tokens that signal non‐completion or withdrawal in sports results (e.g. “dropped out,” “DNF,” “withdrew,” “eliminated”).

    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
     that
    -1.05
     have
    -1.00
     a
    -0.97
     of
    -0.96
     \
    -0.95
     will
    -0.94
     maail
    -0.94
     إذا
    -0.93
     täh
    -0.91
     Hinblick
    -0.91
    POSITIVE LOGITS
     due
    1.09
    rían
    1.09
     noemen
    1.05
    NegativeButton
    1.04
    اسی
    1.02
    ₽
    1.02
     unicornio
    1.00
     ווע
    0.98
     spreken
    0.98
    zedł
    0.97
    Activations Density 0.003%

    No Known Activations