© 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. 10-GEMMASCOPE-RES-131K
    7. 102885
    Prev
    Next
    INDEX
    Explanations

    alphanumeric codes

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

    sequences of digits that form product codes, model numbers, or other technical identifiers.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron detects numeric tokens—that is, sequences of digits (numbers) appearing in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     and
    -2.19
     vervol
    -1.48
     but
    -1.37
     inalámbrico
    -1.36
     Sweatshirt
    -1.35
     Camo
    -1.28
     :—
    -1.24
     Beginnings
    -1.24
    鶺
    -1.22
    必需
    -1.21
    POSITIVE LOGITS
     también
    1.70
     juga
    1.65
     Also
    1.64
    应该
    1.57
     cũng
    1.50
     También
    1.45
     também
    1.43
    有时候
    1.40
     Además
    1.39
     arrêté
    1.37
    Activations Density 0.042%

    No Known Activations