© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2498
    Prev
    Next
    INDEX
    Explanations

    the following concepts

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

    The neuron is detecting numeric literals (especially floating‐point and other numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    tokens that are part of technical code or numerical sequences.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    punctuation-joined tokens in technical text, such as hyphenated terms, dot-separated identifiers/addresses, and contractions.

    oai_token-act-pair · gpt-5Triggered 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
     आरमार
    0.77
     burgundy
    0.74
     réglement
    0.73
     montaña
    0.72
     autumnal
    0.72
     आरमारा
    0.70
     reforzar
    0.69
    हील
    0.69
     tá
    0.68
     Admiralty
    0.68
    POSITIVE LOGITS
    Config
    0.86
    config
    0.82
     config
    0.82
    Default
    0.80
    0.79
    ().
    0.79
    String
    0.78
    Parser
    0.78
    List
    0.78
    初始化
    0.77
    Activations Density 0.008%

    No Known Activations