© 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. 10-GEMMASCOPE-RES-131K
    7. 67898
    Prev
    Next
    INDEX
    Explanations

    gh in URLs and domains

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

    The neuron fires on the letter bigram “gh,” most prominently in strings like the Ghana country‐code top-level domain “.gh.”

    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
    谞
    -2.86
    л
    -2.58
    та
    -2.39
    は
    -2.34
     После
    -2.33
     After
    -2.31
    ↵↵
    -2.28
     With
    -2.20
    With
    -2.19
    し
    -2.19
    POSITIVE LOGITS
    y
    3.89
    r
    2.73
    also
    2.58
    ዣ
    2.39
    ㉖
    2.39
    ープン
    2.34
     hva
    2.34
    毉
    2.31
    眎
    2.31
    うちの
    2.31
    Activations Density 0.007%

    No Known Activations