© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25925
    Prev
    Next
    INDEX
    Explanations

    identifies origin or source

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

    The neuron fires on words that indicate data origin or flow—prepositions and verbs like “from,” “coming,” and “input” that mark where data is coming from.

    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
     spesies
    -0.98
     americano
    -0.95
     fritas
    -0.94
    imwrite
    -0.93
     fero
    -0.90
     Crédito
    -0.90
     tenda
    -0.89
    LookAnd
    -0.89
     vinho
    -0.89
    INDER
    -0.88
    POSITIVE LOGITS
     from
    2.19
     sources
    1.55
    sources
    1.28
     fuentes
    1.27
     Sources
    1.16
     through
    1.16
     través
    1.09
     červená
    1.08
    来源
    1.06
     via
    1.05
    Activations Density 0.103%

    No Known Activations