© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7429
    Prev
    Next
    INDEX
    Explanations

    rule about divisibility

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

    The neuron is signaling how early a token is in the overall document, effectively tracking each token’s relative position from the start.

    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
    os
    0.86
    ين
    0.73
    ско
    0.71
    وس
    0.70
    زا
    0.67
    𝘨
    0.66
    وازي
    0.66
    ِّ
    0.65
    ິ
    0.65
    𝘰
    0.64
    POSITIVE LOGITS
    Myc
    0.77
     obliterated
    0.77
    .'
    0.76
    .\"
    0.75
    ’
    0.74
     совершенно
    0.74
     PDEs
    0.73
     ocurren
    0.73
     которым
    0.71
    高效
    0.71
    Activations Density 0.001%

    No Known Activations