© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38593
    Prev
    Next
    INDEX
    Explanations

    Johnny Wilson Farm LLC

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

    The neuron strongly activates on numeric characters and related punctuation (digits, colons, dashes), i.e. it detects sequences of numbers in raw data or code dumps.

    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
     in
    -2.02
     with
    -1.80
     at
    -1.80
     on
    -1.80
     an
    -1.67
     any
    -1.60
     their
    -1.58
     over
    -1.58
     while
    -1.57
     through
    -1.55
    POSITIVE LOGITS
     decidedly
    1.55
    킁
    1.48
    VERY
    1.48
    actually
    1.45
     unbelievably
    1.41
    烀
    1.41
     Dinas
    1.40
     soooo
    1.38
     REALLY
    1.36
     tamaños
    1.36
    Activations Density 0.001%

    No Known Activations