© 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. 11942
    Prev
    Next
    INDEX
    Explanations

    This neuron fires almost exclusively on the last meaningful content word right before a segment (or “post”) ends, effectively marking end-of-segment boundary tokens.

    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
    legiate
    0.90
     زمانہ
    0.86
    ahami
    0.83
    astudio
    0.80
    زمانہ
    0.79
    ί
    0.79
    ットン
    0.77
    uz
    0.77
    bumptech
    0.76
    uyorum
    0.76
    POSITIVE LOGITS
    с
    0.86
    й
    0.84
    м
    0.82
    та
    0.81
    т
    0.79
    лес
    0.77
    というか
    0.76
    ответ
    0.75
    ле
    0.74
    л
    0.73
    Activations Density 0.000%

    No Known Activations