© 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. 34-GEMMASCOPE-RES-131K
    7. 1735
    Prev
    Next
    INDEX
    Explanations

    technical file headers

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

    The neuron is selectively responding to numeric tokens—digits or numbers embedded 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
     Olympedia
    -0.94
    ТЫ
    -0.90
    vente
    -0.85
    قایناقلار
    -0.85
     ROYAL
    -0.85
    руса
    -0.83
     Sede
    -0.83
    Geograf
    -0.83
     JESUS
    -0.82
     ensalada
    -0.82
    POSITIVE LOGITS
     include
    1.00
     including
    0.82
     we
    0.77
     O
    0.77
     means
    0.72
    })$.
    0.72
    方には
    0.72
     takie
    0.71
    віти
    0.70
     interessieren
    0.70
    Activations Density 0.004%

    No Known Activations