© 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. 22-GEMMASCOPE-RES-131K
    7. 27112
    Prev
    Next
    INDEX
    Explanations

    java classes and technical terms

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

    The neuron fires strongly on class‐declaration lines—especially the inheritance clause (e.g. “extends” or “: BaseClass”) and the opening brace that follows.

    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
     masque
    -0.84
    ᧐
    -0.83
    convite
    -0.81
     reduzir
    -0.80
     设置
    -0.80
     münchen
    -0.78
    Currently
    -0.76
     quietly
    -0.75
     feminina
    -0.75
    Dimensões
    -0.74
    POSITIVE LOGITS
    >"+
    0.90
    jumu
    0.86
    sdk
    0.85
    τει
    0.84
    lorette
    0.82
     momenta
    0.82
    Fatalf
    0.81
    OfType
    0.80
    PANEL
    0.79
    istir
    0.79
    Activations Density 0.006%

    No Known Activations