© 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. 31-GEMMASCOPE-2-RES-16K
    4. 15990
    Prev
    Next
    INDEX
    Explanations

    The neuron spikes on the act-of‐finding verb “find” (and similarly on the coordinating “and” when it’s used to link two main clauses), i.e. it’s detecting that “find”/“and” construction that introduces a location or discovery.

    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
     rivulet
    0.90
    പ്പു
    0.73
    শিরোনাম
    0.72
    رر
    0.70
    ँव
    0.69
     resolução
    0.68
    題
    0.68
     تد
    0.68
    াড়ি
    0.68
    資産
    0.68
    POSITIVE LOGITS
    lington
    0.73
    anken
    0.71
    ych
    0.67
     kam
    0.67
    allerg
    0.65
     poter
    0.65
     поба
    0.65
    aller
    0.64
     Martin
    0.64
    hov
    0.64
    Activations Density 0.000%

    No Known Activations