© 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. 14447
    Prev
    Next
    INDEX
    Explanations

    upvote, status, order, fathers, condition, pos

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

    The neuron is tracking how far into the document you are (i.e. it acts as a positional index of token depth).

    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
    fu
    0.81
     ’
    0.79
    re
    0.77
    zhou
    0.76
     Bach
    0.75
    eid
    0.74
     műkö
    0.73
     %>.
    0.73
    fest
    0.72
     responsabilidad
    0.71
    POSITIVE LOGITS
    ین
    1.05
    ات
    0.90
    ст
    0.82
    тах
    0.82
     véritables
    0.79
    циа
    0.78
    能夠
    0.78
    ವೀ
    0.77
     време
    0.76
    дання
    0.75
    Activations Density 0.000%

    No Known Activations