© 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. 73582
    Prev
    Next
    INDEX
    Explanations

    variable assignment

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

    The neuron fires on all-caps identifier tokens (typically constant or environment-variable names made of uppercase letters and underscores).

    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
     this
    -2.05
     also
    -1.80
     still
    -1.69
     тоже
    -1.58
     by
    -1.58
    :
    -1.54
     might
    -1.45
     seems
    -1.43
     from
    -1.41
     (
    -1.40
    POSITIVE LOGITS
     rése
    1.56
     erila
    1.52
     tärke
    1.52
     besonder
    1.51
     WHILE
    1.49
    ਗ਼
    1.48
     guapos
    1.48
    莇
    1.44
    以外は
    1.42
    伥
    1.42
    Activations Density 0.044%

    No Known Activations