© 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. 54179
    Prev
    Next
    INDEX
    Explanations

    special certain actions

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

    The neuron lights up on marketing‐style product copy, especially on technical or benefit-focused words (ingredients, properties, claims) in promotional descriptions.

    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
    anskje
    -0.96
    ricas
    -0.91
    смотрен
    -0.89
    ariste
    -0.88
     Rhea
    -0.85
     rústica
    -0.85
     fría
    -0.85
     ordentlich
    -0.83
    –>
    -0.83
     วัน
    -0.82
    POSITIVE LOGITS
     certain
    1.45
     ;)
    1.01
     ;-)
    0.97
     that
    0.94
    certain
    0.91
     určit
    0.91
     gewisse
    0.91
     cierta
    0.88
     😉
    0.84
    Certain
    0.84
    Activations Density 0.033%

    No Known Activations