© 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. 10-GEMMASCOPE-RES-131K
    7. 128063
    Prev
    Next
    INDEX
    Explanations

    cannot be acted upon

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

    The neuron detects the phrase “cannot be” (i.e. instances of “cannot be” in the text).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    leicht
    -2.06
     faſt
    -1.94
     内衣
    -1.94
     coloridos
    -1.91
    danke
    -1.91
    aktivi
    -1.86
    schwar
    -1.86
     花纹
    -1.83
    鈇
    -1.82
     sabido
    -1.80
    POSITIVE LOGITS
    on
    2.14
    u
    2.02
    n
    1.99
    ↵
    1.97
    i
    1.94
    .
    1.94
    4
    1.87
    with
    1.76
    When
    1.74
     that
    1.69
    Activations Density 0.023%

    No Known Activations