© 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. 5092
    Prev
    Next
    INDEX
    Explanations

    The neuron is triggered by numeric tokens—especially version numbers, statistics, or other digit‐containing strings.

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

    periods that appear within numeric sequences (especially dates, version numbers, and measurements).

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

    mentions of specific numeric expressions, especially dates, version numbers, and other quantified figures.

    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
    stit
    0.74
    ारात
    0.73
    AN
    0.72
    구요
    0.70
    ANNEL
    0.68
    sta
    0.68
    agl
    0.67
     concentrated
    0.67
    afe
    0.66
     والن
    0.66
    POSITIVE LOGITS
     மாற்ற
    0.76
     dynamism
    0.75
    되어
    0.74
    および
    0.74
     bodyParser
    0.74
     camaraderie
    0.73
     browserSync
    0.72
     aspiring
    0.71
     Snackbar
    0.70
     besondere
    0.70
    Activations Density 0.001%

    No Known Activations