© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 8852
    Prev
    Next
    INDEX
    Explanations

    parent or trusted adult

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

    This neuron is essentially a positional marker—it strongly activates for any token appearing around the 386–420th position in the input sequence.

    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
    ش
    1.05
    یل
    1.04
    ductory
    0.92
    uestros
    0.92
    ם
    0.89
    iguous
    0.89
    مي
    0.88
    هاي
    0.88
    いです
    0.88
    𝗿
    0.88
    POSITIVE LOGITS
     Chakrab
    0.66
     Mr
    0.65
     κε
    0.64
     distaste
    0.64
     largely
    0.63
     COMPLET
    0.63
    Ь
    0.63
    ᥒ
    0.62
     ça
    0.62
     bóng
    0.62
    Activations Density 0.001%

    No Known Activations