© 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. 71901
    Prev
    Next
    INDEX
    Explanations

    questions asking for values

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

    The neuron fires on numeric tokens (individual numbers and numeric punctuation).

    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
     recurrir
    -1.02
    BitConverter
    -0.99
    戋
    -0.96
    ┬
    -0.96
     aload
    -0.94
     подготовка
    -0.94
    nand
    -0.91
    thead
    -0.91
     più
    -0.91
    very
    -0.90
    POSITIVE LOGITS
    こちらも
    1.00
    同じく
    0.96
     any
    0.94
    これも
    0.94
    URBANA
    0.93
    Both
    0.93
     Primaria
    0.92
     quinto
    0.88
     should
    0.88
    Какой
    0.87
    Activations Density 0.010%

    No Known Activations