© 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. 24-GEMMASCOPE-2-RES-16K
    4. 16103
    Prev
    Next
    INDEX
    Explanations

    unittest.TestCase context

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

    The neuron detects parts of proper names and titles (i.e. tokens that belong to named entities).

    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_24_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
    KX
    0.78
    Hr
    0.76
    IN
    0.74
    LZ
    0.74
    E
    0.73
    ON
    0.72
    AT
    0.72
    RO
    0.72
    IS
    0.71
    WALL
    0.70
    POSITIVE LOGITS
    说什么
    0.89
    arthi
    0.85
     حذف
    0.81
    یس
    0.76
     decriminal
    0.75
    فيق
    0.75
     memahami
    0.74
    ്യൂ
    0.73
     PLHIV
    0.73
    ாய்
    0.72
    Activations Density 0.000%

    No Known Activations