© 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. 77820
    Prev
    Next
    INDEX
    Explanations

    technical terms and questions

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

    The neuron activates on code‐style identifiers or signal/parameter names—especially uppercase or all‐caps tokens like POSITION, COLOR, TEXCOORD, CLOCK, WRITE, etc.

    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
    et
    -0.80
    汶
    -0.79
    tle
    -0.78
    又
    -0.78
     again
    -0.77
    有一
    -0.76
    浴
    -0.75
    <h2>
    -0.74
    有人
    -0.73
     Have
    -0.73
    POSITIVE LOGITS
     cães
    1.01
     sementes
    0.94
     cachorros
    0.90
    投资者
    0.83
     espécies
    0.82
    buckle
    0.82
    が欲しい
    0.82
    🔉
    0.82
    ₺
    0.82
    delingen
    0.82
    Activations Density 0.002%

    No Known Activations