© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11593
    Prev
    Next
    INDEX
    Explanations

    mathematical addition

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

    Explanation of neuron 4 behavior: the main thing this neuron does is trigger on nothing—it never activates on any part of the text.

    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
    饱和
    0.91
    )}(\
    0.91
    ਅ
    0.87
    นะครับ
    0.86
    ప్
    0.83
    )}$,
    0.82
    嘭
    0.82
    Ла
    0.81
    तया
    0.81
    ర్
    0.81
    POSITIVE LOGITS
    /−
    1.54
    ++++++++++++++++
    1.14
    ------+
    1.14
    ">+</
    1.09
    &+
    1.05
     (+)
    1.03
    ++++
    0.99
     (++
    0.98
    _+
    0.98
     ++
    0.97
    Activations Density 0.054%

    No Known Activations