© 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. 71616
    Prev
    Next
    INDEX
    Explanations

    technical and formal terms

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

    The neuron strongly activates on title-cased or proper-noun tokens (e.g. names of reports, products, projects, or other domain-specific entities).

    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
    を行い
    -1.52
    きち
    -1.38
     unbelievably
    -1.29
     dessa
    -1.28
    際には
    -1.25
     decidedly
    -1.25
     tremendous
    -1.23
    ふと
    -1.19
     Städten
    -1.19
     incredibly
    -1.19
    POSITIVE LOGITS
     with
    2.11
     in
    2.09
     for
    1.86
     on
    1.58
     by
    1.57
     (
    1.55
     at
    1.44
     without
    1.39
     from
    1.36
     őket
    1.35
    Activations Density 0.036%

    No Known Activations