© 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. 31-GEMMASCOPE-2-RES-16K
    7. 9616
    Prev
    Next
    INDEX
    Explanations

    media size hands clapping

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

    The neuron chiefly responds to numerals and numeric expressions (stand-alone numbers, numbered list markers, ranges, etc.).

    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_31_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
    ––
    0.69
    ά
    0.66
    <unused24>
    0.62
    沭
    0.61
    icmp
    0.61
    мал
    0.60
     verdicts
    0.59
    舌
    0.59
    
    0.59
     –
    0.58
    POSITIVE LOGITS
     —
    1.08
     clapping
    0.96
     clap
    0.95
     ——
    0.82
     Clap
    0.78
     Medium
    0.77
     palms
    0.77
    clap
    0.75
    medium
    0.74
     __
    0.73
    Activations Density 0.135%

    No Known Activations