© 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. 15456
    Prev
    Next
    INDEX
    Explanations

    multi-lingual content

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

    The neuron fires on tokens that look like identifiers or codes—especially any alphanumeric strings containing digits (e.g. version numbers, user‐IDs, dates, or other numeric identifiers).

    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
    oaks
    -1.03
    ᐈ
    -0.99
     trattano
    -0.99
    fourths
    -0.99
    🏡
    -0.98
     Asoci
    -0.97
    zette
    -0.94
     inal
    -0.94
     offensi
    -0.94
    baren
    -0.94
    POSITIVE LOGITS
    both
    1.23
     BOTH
    1.15
    事も
    1.12
    起きた
    1.11
     других
    1.09
    Хорошо
    1.08
     roś
    1.06
    Джере
    1.05
     dueños
    1.04
     deinem
    1.03
    Activations Density 0.038%

    No Known Activations