© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35104
    Prev
    Next
    INDEX
    Explanations

    specific entities and context

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

    The neuron fires on prominent content nouns—especially proper names or specialized technical terms (e.g. “Resort,” “Facebook,” gene or chemical abbreviations, library functions).

    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
     dessa
    -1.23
     שאי
    -1.22
    镚
    -1.19
     incontri
    -1.15
     mencionados
    -1.13
     diretto
    -1.09
     theſe
    -1.08
     şimdi
    -1.03
    помним
    -1.03
    unda
    -1.02
    POSITIVE LOGITS
     Organisator
    1.18
     before
    1.08
     first
    1.06
     Fazit
    1.03
    mə
    1.02
     hauptsächlich
    1.00
    évêque
    1.00
     after
    1.00
     three
    0.99
     during
    0.98
    Activations Density 0.173%

    No Known Activations