© 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. 62901
    Prev
    Next
    INDEX
    Explanations

    dramatic complex situations

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

    The neuron fires on striking, often one-word nouns—mythical, dramatic, or figurative terms that evoke overwhelming or larger-than-life imagery.

    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
     and
    -1.16
    的样子
    -0.96
     exportaciones
    -0.88
    的脸
    -0.86
     muito
    -0.85
    在地上
    -0.84
    能夠
    -0.84
     Highest
    -0.83
    又不是
    -0.82
     fortemente
    -0.82
    POSITIVE LOGITS
     of
    2.48
    般的
    1.88
    like
    1.77
    のような
    1.54
    般
    1.48
    LIKE
    1.46
    esque
    1.43
    zerű
    1.28
     effect
    1.25
    的な
    1.25
    Activations Density 0.069%

    No Known Activations