© 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. 25831
    Prev
    Next
    INDEX
    Explanations

    multiple choice answers

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

    This neuron fires on single-letter answer labels (e.g. “A”, “B”, “C”, etc.) marking chosen multiple‐choice options.

    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
     secrets
    -0.97
    мпу
    -0.96
     について
    -0.94
     Secrets
    -0.92
    オーナー
    -0.90
    secrets
    -0.90
    бр
    -0.90
    娉
    -0.88
     encontré
    -0.88
    ſelf
    -0.87
    POSITIVE LOGITS
    ONAUT
    1.03
     ساتھ
    0.94
    teden
    0.93
     冊
    0.91
    óculos
    0.89
     усі
    0.88
     and
    0.88
     knowled
    0.88
    atitis
    0.88
     رہ
    0.88
    Activations Density 0.002%

    No Known Activations