© 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. 15060
    Prev
    Next
    INDEX
    Explanations

    also followed by descriptive words

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

    The neuron is primarily triggered by verbs (especially contentful past‐ or present‐tense and modal forms).

    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
     autorità
    0.63
    邙
    0.55
    መሳሳይ
    0.55
    ప్రత్య
    0.54
    нали
    0.53
    कोई
    0.52
    ่าง
    0.52
     Remarkably
    0.50
     dacă
    0.50
     Notably
    0.49
    POSITIVE LOGITS
     BUT
    0.62
     pero
    0.56
     lakini
    0.50
     (=
    0.50
     거고
    0.49
     tetapi
    0.48
     lugs
    0.48
    pped
    0.48
     but
    0.47
     কিন্তু
    0.47
    Activations Density 0.038%

    No Known Activations