© 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. 67738
    Prev
    Next
    INDEX
    Explanations

    xor mask, victory lap, gentle use, bed time, panel inspired

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

    This neuron fires on numeric values and technical‐specification terms (e.g. numbers, measurement units, and jargon like “tier,” “deflate,” “xor”).

    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
     gerar
    -1.41
    avril
    -1.36
    そうですが
    -1.32
    性別
    -1.28
     開始
    -1.27
     시작
    -1.27
     необхідно
    -1.27
     націона
    -1.27
     遊戲
    -1.26
    出来ます
    -1.23
    POSITIVE LOGITS
    llten
    1.59
    rrrr
    1.59
     toit
    1.55
    rrrrr
    1.55
    PasswordEncoder
    1.53
    )],
    1.49
    cketed
    1.45
    dienne
    1.43
    )].
    1.43
    riezmann
    1.40
    Activations Density 0.060%

    No Known Activations