© 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. 103885
    Prev
    Next
    INDEX
    Explanations

    comb-related words

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

    The neuron activates on any token containing the character sequence “comb,” regardless of whether it’s a standalone word or part of a larger word.

    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
    .
    -1.78
    -1.73
     dizziness
    -1.68
    ？！”
    -1.55
    b
    -1.49
    -1.48
     voor
    -1.45
    d
    -1.43
    f
    -1.42
     Their
    -1.39
    POSITIVE LOGITS
     włosy
    1.70
    offering
    1.68
     spiele
    1.67
     nascent
    1.66
     .'
    1.66
     penna
    1.65
    --"
    1.59
    grafa
    1.58
    нинг
    1.58
     він
    1.58
    Activations Density 0.010%

    No Known Activations