© 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. 13804
    Prev
    Next
    INDEX
    Explanations

    successful right, cause of, losses among

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

    This neuron fires on factual, content-heavy words—especially technical or domain‐specific nouns and numerical/measured quantities that convey key details.

    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
     Reſ
    -1.20
    ⼆
    -1.16
    戋
    -1.13
     Secretaría
    -1.09
    kkende
    -1.05
    鹇
    -1.05
     unſ
    -1.05
    幛
    -1.00
     juſ
    -0.99
     continually
    -0.98
    POSITIVE LOGITS
     on
    1.07
     vandens
    1.00
     «
    0.96
     european
    0.90
     Although
    0.88
    中でも
    0.87
     aukš
    0.85
     for
    0.84
     formó
    0.82
     barbara
    0.82
    Activations Density 0.112%

    No Known Activations