© 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. 15454
    Prev
    Next
    INDEX
    Explanations

    image acquisition or stream

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

    The neuron primarily fires on English discourse/connective words (e.g. then, again, which) that signal logical or explanatory transitions.

    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
     vários
    -1.09
     gérer
    -1.08
     !!!!
    -1.06
     gedachten
    -1.02
    Ii
    -1.02
    透过
    -1.02
     ‘
    -1.02
    ":
    -1.01
     cérami
    -1.01
     níveis
    -0.99
    POSITIVE LOGITS
     there
    1.65
     because
    1.45
    Viitteet
    1.34
     probably
    1.21
    There
    1.17
     during
    1.17
     before
    1.17
     relatively
    1.16
     intermittently
    1.15
     comparatively
    1.14
    Activations Density 0.097%

    No Known Activations