© 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. 4048
    Prev
    Next
    INDEX
    Explanations

    code blocks or specific instructions

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

    The neuron is looking for explicit quantitative details—numbers, measurements, ages, percentages, and similar numerical or numeric‐style tokens.

    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
     रास्ते
    0.72
     Scope
    0.70
    itarian
    0.66
    प्या
    0.65
     sejahtera
    0.64
    在庫
    0.64
    y
    0.64
    ียร์
    0.63
     sosok
    0.63
    PACK
    0.62
    POSITIVE LOGITS
     гимна
    0.85
    ='
    0.78
    ="
    0.76
    odic
    0.75
    äss
    0.73
     पाच
    0.73
    gateTime
    0.73
    ilem
    0.73
     trots
    0.72
    niego
    0.70
    Activations Density 0.000%

    No Known Activations