© 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. 34-GEMMASCOPE-RES-131K
    4. 109449
    Prev
    Next
    INDEX
    Explanations

    hexadecimal numbers and codes

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

    The neuron activates on digits and numeric punctuation (e.g. individual number tokens and decimal points) in code, citations, or other numeric sequences.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    oinette
    -0.90
     رنگ
    -0.88
     unleash
    -0.82
    🎅
    -0.82
    ்க்க
    -0.82
    👆
    -0.81
    الثة
    -0.81
    */}
    -0.81
     巻
    -0.81
    Kendall
    -0.80
    POSITIVE LOGITS
    complex
    0.84
     his
    0.81
    hili
    0.80
    aré
    0.79
    era
    0.79
    only
    0.79
    étit
    0.78
    orama
    0.77
     ถ
    0.75
    estroyer
    0.75
    Activations Density 0.079%

    No Known Activations