© 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. 33528
    Prev
    Next
    INDEX
    Explanations

    prefixes and code snippets

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

    The neuron fires on multi‐letter sequences that are written in title case or all‐caps (i.e. substrings of identifiers or proper names that start with an uppercase letter).

    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
     adopté
    -1.24
     disant
    -1.17
     payé
    -1.16
     herói
    -1.16
    榦
    -1.15
    vily
    -1.13
     hâte
    -1.13
     pronon
    -1.12
    ouncements
    -1.12
    où
    -1.12
    POSITIVE LOGITS
     while
    1.34
     اكتوبر
    1.14
     when
    1.05
    してみると
    1.04
     "-";
    1.03
     renegade
    1.00
     this
    1.00
     then
    0.97
     '-';
    0.97
     which
    0.96
    Activations Density 0.057%

    No Known Activations