© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13815
    Prev
    Next
    INDEX
    Explanations

    code, punctuation, and numbers

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

    The main thing this neuron does is detect floating-point numbers (numeric tokens containing a decimal point).

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

    numeric digits, particularly in dates and sequences.

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

    explicit references to time, especially numerals in dates and other temporal expressions within sentences.

    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_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
    秞
    0.93
     अवेलेबल
    0.92
     pertenec
    0.91
    壺
    0.90
     ತಯ
    0.86
    िरिक्स
    0.84
    িতি
    0.83
     ete
    0.81
     panggilan
    0.80
     bhave
    0.80
    POSITIVE LOGITS
    fur
    0.80
    0.77
    ager
    0.76
    wolves
    0.72
    )
    0.72
    ob
    0.70
    gra
    0.70
    pins
    0.68
    3
    0.68
    ↵
    0.68
    Activations Density 0.000%

    No Known Activations