© 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. 10-GEMMASCOPE-RES-131K
    7. 65405
    Prev
    Next
    INDEX
    Explanations

    information and its source

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

    The neuron fires on occurrences of the noun “information,” especially when it’s used in phrases describing information being contained, given, sent, supplied, etc.

    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
    OGUE
    -1.63
    𝗔
    -1.63
    --;
    -1.62
    ENEZ
    -1.60
    backslash
    -1.58
    朦
    -1.52
    数十
    -1.50
     перспекти
    -1.48
     می‌ده
    -1.48
    音乐
    -1.47
    POSITIVE LOGITS
     about
    1.85
    jeuner
    1.64
     information
    1.62
     from
    1.59
    information
    1.54
    といふ
    1.54
     camadas
    1.52
    .
    1.52
     remains
    1.52
     كان
    1.50
    Activations Density 0.119%

    No Known Activations