© 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. 22-GEMMASCOPE-RES-131K
    7. 33363
    Prev
    Next
    INDEX
    Explanations

    fetching all

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

    This neuron fires on code tokens that name or invoke list‐oriented retrieval methods (e.g. “index,” “all,” “get…,” “find…All”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     with
    -1.22
     when
    -1.11
     caused
    -1.06
     after
    -1.00
    щается
    -0.98
    NSIndexPath
    -0.95
    чений
    -0.93
    高峰
    -0.92
     in
    -0.91
    和
    -0.90
    POSITIVE LOGITS
    umerable
    1.20
     débuts
    1.17
    Soorten
    1.16
    Who
    1.13
     Институ
    1.11
    一个个
    1.10
    Aktuelle
    1.09
    lovely
    1.08
    latest
    1.05
     FORCES
    1.05
    Activations Density 0.009%

    No Known Activations