© 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. 14017
    Prev
    Next
    INDEX
    Explanations

    Let X =

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

    The neuron fires on phrases that introduce or define mathematical objects—e.g. “choose x₁ = a,” “define a continuous function f on [0,1] by f(x)=a,” etc.

    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
     bolsillo
    -1.00
     sowohl
    -0.98
     både
    -0.96
     zowel
    -0.94
     geöffnet
    -0.94
    \'
    -0.93
    }$-
    -0.91
     różnych
    -0.91
     específicas
    -0.91
     могло
    -0.91
    POSITIVE LOGITS
    ῷ
    0.96
     solidari
    0.93
     增加
    0.89
    就可以了
    0.89
    lieving
    0.88
     діяльності
    0.88
     whatever
    0.86
     any
    0.86
    唱
    0.85
    出了一
    0.85
    Activations Density 0.024%

    No Known Activations