© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66706
    Prev
    Next
    INDEX
    Explanations

    legal names, citations, and code variables

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

    The neuron activates on identifier tokens—especially variable or property names—within code (e.g. sr, checkWasCalled, txtCode, this.name, variables, MAKE_FLAGS, fileCounter).

    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
    initialState
    -0.97
     中央
    -0.90
     liest
    -0.88
     meste
    -0.88
    ৷
    -0.86
    法师
    -0.86
     predominant
    -0.85
    typeOf
    -0.84
     Scott
    -0.83
     turut
    -0.82
    POSITIVE LOGITS
     estrenar
    0.94
     گوید
    0.90
     leçon
    0.90
     outubro
    0.88
    mover
    0.86
    Dimensões
    0.86
    ерез
    0.84
     totalité
    0.84
     plegable
    0.83
     Koning
    0.83
    Activations Density 0.004%

    No Known Activations