© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 50088
    Prev
    Next
    INDEX
    Explanations

    permission READ external_storage

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

    The neuron specializes in spotting numeric tokens (strings composed of digits).

    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
    暑假
    -0.77
    arent
    -0.69
    Scalar
    -0.69
    帳に追加
    -0.68
     dates
    -0.65
     aceite
    -0.64
     возрасте
    -0.63
     Cf
    -0.63
     والك
    -0.63
    ücke
    -0.63
    POSITIVE LOGITS
     Judith
    0.71
    워
    0.67
    Judith
    0.67
    Kota
    0.66
     Nay
    0.65
     ciò
    0.64
    ierend
    0.62
     ferrous
    0.62
    Nay
    0.62
    群馬
    0.61
    Activations Density 0.071%

    No Known Activations