© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8752
    Prev
    Next
    INDEX
    Explanations

    1

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

    The neuron strongly activates on numeric tokens (claim step numbers and punctuation) used to enumerate sections or clauses in patent-style documents.

    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_31_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
     друзья
    0.62
    )
    0.59
    நாயகம்
    0.57
     camisetas
    0.56
     ঈশ্বরের
    0.55
    朋友
    0.55
     друзей
    0.54
     terapé
    0.53
    efeu
    0.52
     plaques
    0.52
    POSITIVE LOGITS
     \(
    0.77
    \.
    0.63
    \(
    0.60
    、​
    0.60
    killed
    0.59
     என்
    0.57
    killing
    0.56
    round
    0.56
    item
    0.55
    \)
    0.55
    Activations Density 0.033%

    No Known Activations