© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 9557
    Prev
    Next
    INDEX
    Explanations

    boolean values and hex codes

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

    The neuron is detecting numeric literals (decimal or hexadecimal number tokens) in code.

    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
     during
    -1.67
     after
    -1.63
     with
    -1.49
    -
    -1.42
     through
    -1.38
     by
    -1.34
     or
    -1.34
     within
    -1.34
     (
    -1.33
     which
    -1.33
    POSITIVE LOGITS
     joyous
    1.82
     joyful
    1.69
     magnificent
    1.56
     astounding
    1.55
     tumultuous
    1.55
     majestic
    1.53
     horrifying
    1.47
     astonishing
    1.45
     sumptuous
    1.45
     gorgeous
    1.40
    Activations Density 0.001%

    No Known Activations