© 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. 73465
    Prev
    Next
    INDEX
    Explanations

    numbers, prefixes, names, code

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

    The neuron primarily detects numeric tokens and figures (e.g. standalone numbers, percentages, and numerals).

    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
     migra
    -0.86
     Alexa
    -0.86
     obed
    -0.85
    További
    -0.82
    Alexa
    -0.82
     tasche
    -0.82
    EMY
    -0.81
     MEX
    -0.80
    rolid
    -0.80
     herbes
    -0.80
    POSITIVE LOGITS
     groter
    1.04
     ostavi
    0.98
    ではありません
    0.93
     tax
    0.93
    使用する
    0.91
     daugiau
    0.89
    nect
    0.88
    ilier
    0.88
    渽
    0.88
     krę
    0.85
    Activations Density 0.001%

    No Known Activations