© 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. 40330
    Prev
    Next
    INDEX
    Explanations

    types of things

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

    The neuron reliably flags content-bearing, often multi‐syllabic nouns and technical terms (rather than 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
     was
    -0.97
    glement
    -0.94
     autre
    -0.93
     relacionado
    -0.93
    anor
    -0.90
     contexto
    -0.90
     mapStateToProps
    -0.89
    了一眼
    -0.89
     descripción
    -0.88
     custody
    -0.88
    POSITIVE LOGITS
     you
    1.42
     we
    1.42
     that
    1.38
     available
    1.08
    asiest
    1.05
     guatemala
    1.02
     who
    0.99
     jesteśmy
    0.98
     necessary
    0.98
    wonderful
    0.97
    Activations Density 0.112%

    No Known Activations