© 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. 128085
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect numeric tokens (numbers and numeric expressions) in technical text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the associated matrix

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     winning
    -0.92
     fatal
    -0.90
    </b>
    -0.86
    7
    -0.86
     дене
    -0.86
    有三
    -0.86
     играет
    -0.85
     komple
    -0.85
     dokumenti
    -0.85
     TOO
    -0.85
    POSITIVE LOGITS
     that
    1.33
     marcadas
    0.98
    UrlEncoded
    0.94
    ̬
    0.93
     mencer
    0.92
    広
    0.91
    いま
    0.90
     pantalón
    0.90
    抡
    0.88
     personalizadas
    0.85
    Activations Density 0.088%

    No Known Activations