© 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. 27712
    Prev
    Next
    INDEX
    Explanations

    actions and responses

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

    The main thing this neuron does is pick out unusually rare or low-frequency tokens (words or symbols).

    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
     other
    -0.96
     several
    -0.88
     Brown
    -0.85
     demás
    -0.83
     chose
    -0.82
     many
    -0.81
     sofort
    -0.81
     Naturs
    -0.80
    ラボ
    -0.79
    findpost
    -0.77
    POSITIVE LOGITS
     respons
    0.96
     theirs
    0.93
    Respons
    0.88
    ACCEPT
    0.87
     پاسخ
    0.86
    홈
    0.85
     asti
    0.84
    Demografie
    0.82
     cordless
    0.82
     Respons
    0.81
    Activations Density 0.065%

    No Known Activations