© 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. 156
    Prev
    Next
    INDEX
    Explanations

    flash messages and alerts

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

    The neuron fires strongly on numeric tokens (integer literals or standalone numbers) 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
    éda
    -0.88
    ülő
    -0.85
    ENSOR
    -0.84
    やかに
    -0.83
     impover
    -0.81
    ⅲ
    -0.80
    かず
    -0.80
    UST
    -0.79
     is
    -0.79
    に住
    -0.79
    POSITIVE LOGITS
     успе
    1.10
     successful
    1.09
     success
    1.09
     successfully
    1.05
     achieved
    0.95
    successful
    0.94
     成功
    0.93
     message
    0.92
     successes
    0.90
    Toolkit
    0.88
    Activations Density 0.008%

    No Known Activations