© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130464
    Prev
    Next
    INDEX
    Explanations

    something new or different

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

    The main thing this neuron does is detect occurrences of the word “new.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    秾
    -1.88
    helps
    -1.88
     maksi
    -1.81
    咻
    -1.80
     énorme
    -1.79
     kitu
    -1.77
     pembe
    -1.75
     帅气
    -1.75
    錼
    -1.73
    ほら
    -1.72
    POSITIVE LOGITS
     bądź
    1.80
    ."
    1.72
     inicios
    1.66
    %\]
    1.66
     incl
    1.66
    6
    1.65
     or
    1.59
     dismiss
    1.55
    ку
    1.55
     they
    1.55
    Activations Density 0.008%

    No Known Activations