© 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. 22-GEMMASCOPE-RES-131K
    7. 15698
    Prev
    Next
    INDEX
    Explanations

    hexadecimal numbers and colors

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

    The neuron activates on occurrences of “hex,” especially when used to denote hexadecimal values or functions (e.g. hex‐to‐decimal, hex strings, hex colors).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ”
    -2.64
    ↵↵
    -2.61
    }
    -2.56
     encompassed
    -2.22
    縂
    -2.17
     cuando
    -2.16
    ,”
    -2.16
     essas
    -2.14
    盌
    -2.09
    .”
    -2.06
    POSITIVE LOGITS
    IFYING
    2.47
    徘
    2.14
     Ре
    2.08
    ſtra
    2.03
     \#
    1.99
    沺
    1.98
    꽤
    1.98
    蜮
    1.98
     fondness
    1.95
    搦
    1.95
    Activations Density 0.010%

    No Known Activations