© 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. 2420
    Prev
    Next
    INDEX
    Explanations

    6 or

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

    The neuron is triggered by long sequences of digits (e.g. phone numbers or other multi‐digit numeric strings).

    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
     progenitors
    0.63
    करों
    0.60
     ดังนั้น
    0.57
     adsorbed
    0.57
     reborn
    0.57
     ternyata
    0.56
    蘖
    0.56
     trotz
    0.55
     malgré
    0.55
     ibid
    0.55
    POSITIVE LOGITS
     eller
    0.83
     અથવા
    0.83
     hoặc
    0.82
    или
    0.82
     helpline
    0.82
    하거나
    0.81
     или
    0.76
    或者
    0.75
    📞
    0.75
     atau
    0.75
    Activations Density 0.028%

    No Known Activations