© 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. 47736
    Prev
    Next
    INDEX
    Explanations

    mathematical and artistic queries

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

    The neuron fires on lines composed of long runs of punctuation or ASCII‐art border characters (e.g. repeated “*”, “–”, “/”, “\”, etc.) rather than on normal words.

    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
    衫
    -0.98
     daardoor
    -0.91
     бампер
    -0.90
    וּ
    -0.87
    Position
    -0.84
     excessively
    -0.83
    ̫
    -0.83
     or
    -0.83
    タッチ
    -0.82
    ởi
    -0.81
    POSITIVE LOGITS
    nemia
    0.97
    襖
    0.96
    嘴角
    0.91
     caminh
    0.89
     /\
    0.88
     Nähr
    0.87
     nust
    0.86
     meninas
    0.86
    gäng
    0.86
    了一个
    0.85
    Activations Density 0.031%

    No Known Activations