© 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. 34-GEMMASCOPE-RES-131K
    7. 25826
    Prev
    Next
    INDEX
    Explanations

    initiating statement or query

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

    The neuron is essentially a “number detector,” turning on most strongly whenever it encounters a numeric token 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_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
     hecho
    -0.96
     игрушки
    -0.90
     febrero
    -0.84
     mulher
    -0.81
    getColumnIndex
    -0.80
     medizin
    -0.80
    万年
    -0.80
    ál
    -0.80
     cinco
    -0.79
    ября
    -0.79
    POSITIVE LOGITS
     allerlei
    0.94
    atelli
    0.89
     yaitu
    0.88
    Sadly
    0.87
     Очень
    0.85
    Lots
    0.85
     fume
    0.84
     очень
    0.84
     anglo
    0.84
    lots
    0.83
    Activations Density 0.000%

    No Known Activations