© 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. 2331
    Prev
    Next
    INDEX
    Explanations

    This neuron indiscriminately activates on ordinary word tokens (letters forming words) as opposed to punctuation, whitespace, or non-word symbols.

    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
    کہ
    0.94
    ש
    0.79
    𝗮
    0.79
    attiyam
    0.75
    ਨੇ
    0.75
    ства
    0.75
    ประเภท
    0.74
    სახ
    0.73
     quintessential
    0.72
    କ
    0.71
    POSITIVE LOGITS
     Inglaterra
    0.73
    တင်း
    0.69
    umble
    0.69
     estaban
    0.68
     invitados
    0.68
     وعلى
    0.67
    yce
    0.67
     Pain
    0.66
    ಾಗುತ್ತದೆ
    0.66
     Coroner
    0.66
    Activations Density 0.000%

    No Known Activations