© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130731
    Prev
    Next
    INDEX
    Explanations

    The neuron is primarily triggered by numeric tokens—especially floating‐point numbers (i.e. tokens containing a decimal point).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    factor of

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    に挑戦
    -1.03
    でしたね
    -1.00
     testing
    -0.97
     rám
    -0.97
     filós
    -0.96
     permettra
    -0.95
    になりますが
    -0.93
     avanzado
    -0.91
    näm
    -0.91
    jetivo
    -0.90
    POSITIVE LOGITS
    ifes
    0.85
     etc
    0.84
    ạp
    0.82
     peringkat
    0.79
    ITECH
    0.79
     ambao
    0.78
     recollections
    0.78
     ''
    
    0.77
    ances
    0.77
     inferior
    0.76
    Activations Density 0.001%

    No Known Activations