© 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. 14595
    Prev
    Next
    INDEX
    Explanations

    bless their hearts

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

    This neuron fires strongly on the subordinating conjunction “while” (including the variant “whilst”) when it introduces a clause.

    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
    א
    0.79
     প্রমাণিত
    0.71
    وضع
    0.70
     hadrons
    0.70
     psychiatrist
    0.68
    adan
    0.67
    ولة
    0.66
    bagai
    0.65
     থাকা
    0.65
    ين
    0.65
    POSITIVE LOGITS
    鞦
    0.82
     twelfth
    0.82
     Déc
    0.81
    یی
    0.79
     Nee
    0.78
    经
    0.78
    heartedly
    0.77
    亾
    0.77
    秶
    0.75
     conventionally
    0.74
    Activations Density 0.000%

    No Known Activations