© 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. 13376
    Prev
    Next
    INDEX
    Explanations

    1 point

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

    The neuron is highly responsive to numeric tokens and numeric expressions (e.g. digits, chapter or section numbers, quantities).

    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
     envisaged
    0.70
     advices
    0.68
     applications
    0.62
    ↵↵
    0.61
     Anand
    0.59
     cheap
    0.58
     supplying
    0.57
     coloured
    0.56
     对
    0.56
     equipments
    0.55
    POSITIVE LOGITS
     "¿
    0.76
     estadounidenses
    0.75
    Ꮺ
    0.73
    usura
    0.72
     liderazgo
    0.69
     gerente
    0.67
     ahorita
    0.67
    วรร
    0.67
    Capstone
    0.67
    Ꮦ
    0.67
    Activations Density 0.140%

    No Known Activations