© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 58379
    Prev
    Next
    INDEX
    Explanations

    license terms and copyright

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

    The neuron is primarily sensitive to numeric tokens (e.g. literal numbers and number‐related symbols).

    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
     viktigt
    -0.86
    øya
    -0.84
    وع
    -0.81
    rägen
    -0.79
     maç
    -0.77
    regions
    -0.77
     luxus
    -0.77
     akty
    -0.77
     kabát
    -0.77
    initializer
    -0.77
    POSITIVE LOGITS
     nhập
    0.85
     entry
    0.83
     imports
    0.83
     дитя
    0.75
    entry
    0.74
     COLUMN
    0.73
    diak
    0.72
    每个人
    0.72
    testen
    0.71
    xically
    0.71
    Activations Density 0.003%

    No Known Activations