© 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. 34-GEMMASCOPE-RES-131K
    7. 116981
    Prev
    Next
    INDEX
    Explanations

    doing a task

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

    The neuron consistently fires on numerical tokens—digits, version numbers, constants, and other number-like values 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
    ֮
    -0.85
    CAREER
    -0.79
    cuid
    -0.74
     początku
    -0.72
     ogrom
    -0.72
     cuci
    -0.71
    Tanya
    -0.71
    Unset
    -0.70
    getDb
    -0.70
     穴
    -0.70
    POSITIVE LOGITS
     task
    1.84
     Task
    1.54
    Task
    1.52
     tasks
    1.50
    task
    1.37
    任务
    1.29
     Tasks
    1.18
     tick
    1.16
    Tasks
    1.15
     TASK
    1.04
    Activations Density 0.016%

    No Known Activations