© 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. 54254
    Prev
    Next
    INDEX
    Explanations

    appreciate effort and skill

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

    The neuron is a general “content‐word” detector, firing strongly on informative nouns, verbs, adjectives, and adverbs (i.e. non–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
    QtWidgets
    -1.04
     étrange
    -1.01
    踞
    -0.95
    ɵɵ
    -0.93
     zeer
    -0.92
    prite
    -0.92
     আম
    -0.92
    カシ
    -0.92
     Gewebe
    -0.92
     jungen
    -0.91
    POSITIVE LOGITS
     appreciation
    1.23
     enjoyment
    1.09
     TÉCNICA
    1.05
     visuelle
    1.05
     tages
    1.05
    ׇ
    1.04
     deserving
    1.03
     worth
    1.03
    fabri
    0.99
     appreciate
    0.95
    Activations Density 0.034%

    No Known Activations