© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 104349
    Prev
    Next
    INDEX
    Explanations

    biological and social concepts

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

    academic and scientific terminology related to measurements, effects, or processes.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron preferentially activates on unusually long tokens (multi‐syllabic words or long word fragments).

    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
     longterm
    -0.93
    ‌.
    -0.91
    findFirst
    -0.87
    เพื่อ
    -0.85
     her
    -0.85
     lends
    -0.83
    reathable
    -0.82
     to
    -0.82
     дитя
    -0.81
     sixth
    -0.79
    POSITIVE LOGITS
    ümü
    0.92
    uos
    0.92
     واتس
    0.91
    lleri
    0.89
    ROOT
    0.89
    ulkan
    0.88
     amitié
    0.85
    рки
    0.85
    liko
    0.84
     посред
    0.83
    Activations Density 0.049%

    No Known Activations