© 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. 34-GEMMASCOPE-RES-131K
    7. 57936
    Prev
    Next
    INDEX
    Explanations

    names, places, or things

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

    The neuron fires on numeric tokens (years, event numbers, scores) and on tokens that label or title those numeric references (e.g. film years, UFC event numbers, championship dates).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     COVID
    -0.81
     Thickness
    -0.79
     Method
    -0.79
     Affect
    -0.76
    zyn
    -0.76
    🧰
    -0.72
     Bess
    -0.71
    观点
    -0.71
    sistency
    -0.70
     flung
    -0.70
    POSITIVE LOGITS
     Darfur
    0.70
     Dac
    0.70
     Waw
    0.69
     MySpace
    0.68
    ൾ
    0.67
     Sequ
    0.67
     Inuit
    0.65
    Regen
    0.65
    Scarlett
    0.65
    informa
    0.64
    Activations Density 0.038%

    No Known Activations