© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 8006
    Prev
    Next
    INDEX
    Explanations

    task or task-related nouns

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

    This neuron fires on prominent headings or titles—especially workshop names, section headers, and other all-caps or stylized title phrases.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Polyg
    0.61
     incest
    0.61
     geoLocation
    0.55
    ಹು
    0.54
     Bragg
    0.54
     thương
    0.53
    सीसीआई
    0.53
    嬰
    0.53
    emper
    0.53
     tumeurs
    0.52
    POSITIVE LOGITS
     tasks
    1.95
    任务
    1.84
    tasks
    1.82
    Tasks
    1.80
     Tasks
    1.80
     task
    1.74
     tarefas
    1.66
    task
    1.63
     tareas
    1.60
     deadlines
    1.59
    Activations Density 0.750%

    No Known Activations