© 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. 20673
    Prev
    Next
    INDEX
    Explanations

    programming frame

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

    The neuron strongly activates on occurrences of the token “frame” (e.g. as a JSON key or code identifier). It’s essentially looking for the word “frame.”

    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
     แปล
    -1.76
     买
    -1.68
     einzigen
    -1.66
     缝
    -1.65
    ƺ
    -1.64
    abung
    -1.63
     绘画
    -1.61
     ภาพถ่าย
    -1.60
    Reminds
    -1.59
    狁
    -1.59
    POSITIVE LOGITS
    man
    1.80
    centerTitle
    1.75
    (
    1.73
    ?
    1.73
     ambiance
    1.73
     Uncategorized
    1.71
    ازات
    1.70
    cessor
    1.68
    渽
    1.66
    longrightarrow
    1.63
    Activations Density 0.011%

    No Known Activations