© 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. 6715
    Prev
    Next
    INDEX
    Explanations

    node structure, command, result, seen

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

    The neuron is acting as a crude “document‐position” detector, firing most strongly on tokens near the start of the text and then gradually decaying as you move further into the document.

    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
    σταν
    1.04
     chond
    0.96
     Teile
    0.93
     hedon
    0.93
     potenti
    0.91
     Tafel
    0.91
     Begriff
    0.91
     batal
    0.89
     stellen
    0.88
     roten
    0.88
    POSITIVE LOGITS
    REAL
    0.86
    Reducing
    0.76
    Protect
    0.74
    Painted
    0.73
    Algun
    0.71
    uc
    0.70
    뮤
    0.70
    Already
    0.70
    사랑
    0.70
    Phys
    0.69
    Activations Density 0.000%

    No Known Activations