© 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. 16353
    Prev
    Next
    INDEX
    Explanations

    anything or everything

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

    The neuron activates on topic-bearing content words—i.e. relatively rare, domain-specific nouns and verbs conveying the core subject matter.

    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
     беже
    -0.92
     inexhaustible
    -0.90
    uvu
    -0.86
     operation
    -0.85
    Those
    -0.85
    bilis
    -0.85
     Enough
    -0.85
    /*",
    -0.84
     More
    -0.83
     Keep
    -0.81
    POSITIVE LOGITS
     anything
    1.32
    various
    1.28
    anything
    1.28
    everything
    1.27
     everything
    1.16
     ANYTHING
    1.16
     apapun
    1.14
     انواع
    1.13
     berbagai
    1.12
     EVERYTHING
    1.11
    Activations Density 0.187%

    No Known Activations