© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 563
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on self-referential language and directive verbs—especially first-person pronouns (“I,” “my,” “somebody”) and modals or imperatives (“need,” “to,” “do,” “say”).

    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
     
    0.69
    :
    0.68
    (
    0.66
    ↵↵
    0.61
    :
    0.60
    。
    0.60
    /
    0.58
    +
    0.54
    市
    0.52
    {
    0.52
    POSITIVE LOGITS
     وړاندوینې
    0.75
     یہ
    0.75
    ټبال
    0.75
     بیاکت
    0.74
     pięk
    0.73
     schönsten
    0.73
     ""`
    0.73
     څرنګ
    0.72
     sarebbe
    0.72
     crianças
    0.70
    Activations Density 0.000%

    No Known Activations