© 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. 9228
    Prev
    Next
    INDEX
    Explanations

    null, None, nullptr, nil checks

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

    The neuron fires on numerical tokens (integers, decimals, IDs, version numbers, etc.) in 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
     purely
    0.60
    ተት
    0.59
    every
    0.58
     mathematically
    0.57
    厳
    0.57
    ধে
    0.57
     احنا
    0.55
    EVERY
    0.54
    백
    0.54
     mathematical
    0.54
    POSITIVE LOGITS
     null
    3.11
    null
    2.89
     NULL
    2.63
    NULL
    2.59
     nullptr
    2.34
    nullptr
    2.20
    Null
    2.19
     Null
    2.19
     nul
    1.90
     nil
    1.83
    Activations Density 0.033%

    No Known Activations