© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13152
    Prev
    Next
    INDEX
    Explanations

    age calculation

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

    The neuron is essentially flagging “out‐of‐distribution” or anomalous tokens—i.e. words that appear in a different style or domain than the surrounding 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_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
    т
    0.88
     ר
    0.83
     passato
    0.81
    И
    0.81
    ک
    0.76
     funzione
    0.76
     lumière
    0.75
     மிகப்பெரிய
    0.75
    Ş
    0.74
     д
    0.73
    POSITIVE LOGITS
     Reconciliation
    0.82
     JANUARY
    0.77
     POLIT
    0.76
    жая
    0.74
    なります
    0.72
     Rober
    0.71
     Courtesy
    0.71
     Tilly
    0.71
     OZ
    0.71
    苗
    0.71
    Activations Density 0.000%

    No Known Activations