© 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. 28191
    Prev
    Next
    INDEX
    Explanations

    neighbour, order, type

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

    The neuron activates on numeric tokens—sequences of digits (including those with decimal points) in the text.

    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
    -1.26
     where
    -0.93
     África
    -0.89
     Very
    -0.86
     on
    -0.85
     made
    -0.85
     got
    -0.85
     now
    -0.85
     all
    -0.84
     when
    -0.84
    POSITIVE LOGITS
     presunto
    0.93
    ほとん
    0.93
     auseinander
    0.91
     gelegt
    0.90
     vakantie
    0.90
     reager
    0.90
    儸
    0.89
     jesteśmy
    0.88
     hunne
    0.87
    蓠
    0.87
    Activations Density 0.021%

    No Known Activations