© 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. 7442
    Prev
    Next
    INDEX
    Explanations

    compiled program or executable

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

    The neuron is especially sensitive to numeric tokens (years, page or volume numbers, and decimal values).

    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
    ्स
    0.82
    ammatory
    0.75
    nominated
    0.71
    tooltip
    0.70
    ͆
    0.70
    了不少
    0.69
    singleton
    0.69
    theless
    0.68
    वाची
    0.68
    诠
    0.68
    POSITIVE LOGITS
     Fase
    0.85
     Shazam
    0.84
     Disk
    0.83
     Fass
    0.81
     Clothes
    0.79
     Cortes
    0.79
     Faites
    0.79
     Để
    0.78
    Disk
    0.76
     Forman
    0.75
    Activations Density 0.001%

    No Known Activations