© 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. 65158
    Prev
    Next
    INDEX
    Explanations

    initiating after time

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

    This neuron primarily fires on common function words and simple grammatical connectors (e.g. conjunctions, prepositions, temporal markers) rather than on content-bearing vocabulary.

    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
     getrennt
    -0.95
     върху
    -0.94
     renseignements
    -0.94
    いま
    -0.94
     …
    -0.91
    ॑
    -0.91
     refuer
    -0.90
     большинство
    -0.89
     която
    -0.88
     والتي
    -0.87
    POSITIVE LOGITS
     профилакти
    1.09
     fernández
    1.08
    裨
    0.99
     тера
    0.97
     filtre
    0.97
     tâ
    0.97
    ⸒
    0.97
     momentos
    0.96
     onResume
    0.96
    龶
    0.95
    Activations Density 0.089%

    No Known Activations