© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9338
    Prev
    Next
    INDEX
    Explanations

    argument exception local ions

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

    The neuron is essentially looking for words that occur as part of quantitative specifications—i.e. the unit‐ or measurement-bearing tokens (durations, dosages, quantitative prefixes/suffixes) that appear adjacent to numbers.

    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
    ousand
    0.88
    ાન
    0.85
     家庭
    0.82
    茹
    0.79
     พอ
    0.78
    鬢
    0.77
     編
    0.77
     カーボン
    0.76
     મંદિર
    0.76
    勬
    0.75
    POSITIVE LOGITS
     cuyo
    0.77
     técn
    0.73
     asociado
    0.73
     merupakan
    0.71
     asociada
    0.71
     buscan
    0.71
    ('-',
    0.71
    ffen
    0.70
     berasal
    0.70
     terdiri
    0.70
    Activations Density 0.000%

    No Known Activations