© 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. 76802
    Prev
    Next
    INDEX
    Explanations

    Brigham Young, Vikram Bhatt

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

    The neuron strongly activates on tokens containing the character sequence “ram” (e.g. Brigham, diagrammatic, Abramowitz, Vikram).

    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
    gól
    -0.96
    個人的
    -0.96
     homes
    -0.93
    anatom
    -0.91
    ائمة
    -0.90
     months
    -0.88
     który
    -0.88
    رحلة
    -0.84
    pont
    -0.82
     lament
    -0.81
    POSITIVE LOGITS
    matic
    1.22
    icity
    1.09
    mes
    1.00
    mer
    0.93
    med
    0.93
    cto
    0.91
    MING
    0.90
     ヤ
    0.85
    ming
    0.84
    ite
    0.84
    Activations Density 0.048%

    No Known Activations