© 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. 34-GEMMASCOPE-RES-131K
    7. 31564
    Prev
    Next
    INDEX
    Explanations

    stock tickers

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

    The neuron flags tokens that are uppercase strings or numerals—e.g. acronyms, identifiers, error-codes, and other all-caps or number-containing terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Kälte
    -0.84
     mueve
    -0.83
     moeilijk
    -0.76
    占用
    -0.73
    ,''
    -0.72
    IMPORTED
    -0.72
     créations
    -0.72
    дайте
    -0.71
    nextDouble
    -0.71
    ODD
    -0.71
    POSITIVE LOGITS
     NZ
    0.88
    ACKNOWLEDG
    0.86
    ymbol
    0.85
     ticker
    0.85
     PY
    0.83
     INSTRUMENTS
    0.81
     KT
    0.81
    符号
    0.80
    стоит
    0.80
    privileged
    0.79
    Activations Density 0.213%

    No Known Activations