© 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. 22-GEMMASCOPE-RES-131K
    7. 98276
    Prev
    Next
    INDEX
    Explanations

    introduces descriptive phrases

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

    The neuron activates on numeric and quantitative detail tokens (e.g. numbers, units, ratings, distances, times) in descriptive 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_22/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
     required
    -0.97
     require
    -0.93
     its
    -0.91
    ضل
    -0.90
     sozialen
    -0.89
    用途
    -0.86
     renal
    -0.83
    要不是
    -0.82
    gridColumn
    -0.82
     requires
    -0.82
    POSITIVE LOGITS
     Assay
    0.94
    тили
    0.92
     vectoriales
    0.92
    AKA
    0.91
     Ionic
    0.91
     peruan
    0.91
     εδώ
    0.90
     pollo
    0.88
     ack
    0.88
     ショルダー
    0.87
    Activations Density 0.003%

    No Known Activations