© 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. 72677
    Prev
    Next
    INDEX
    Explanations

    public files and records

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

    The neuron consistently lights up on content-bearing words (longer nouns, verbs, and technical identifiers) rather than on short function words.

    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
     Perceived
    -0.95
    CDT
    -0.87
     superimposed
    -0.79
     POEMS
    -0.79
    再次
    -0.78
     obrazu
    -0.77
    出場
    -0.75
    SHAW
    -0.75
     configuración
    -0.74
     입력
    -0.74
    POSITIVE LOGITS
     accessible
    0.97
    public
    0.96
    ractable
    0.93
     public
    0.92
    旗下
    0.86
     detectable
    0.86
     their
    0.83
     寫
    0.81
     Public
    0.80
    лков
    0.80
    Activations Density 0.176%

    No Known Activations