© 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. 15982
    Prev
    Next
    INDEX
    Explanations

    metal music related terms

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

    This neuron responds to the opening and closing of direct‐speech quotes (i.e. it marks segments of text enclosed in quotation marks).

    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
     delightful
    0.64
     bhaj
    0.63
     ♪
    0.62
     jazz
    0.61
     rnb
    0.60
     pac
    0.59
     bach
    0.57
    achtree
    0.57
     opal
    0.57
     remix
    0.55
    POSITIVE LOGITS
    Metal
    1.24
     METAL
    1.23
     Metal
    1.20
    METAL
    1.18
    🤘
    1.13
    metal
    1.12
     metal
    1.06
    金属
    1.04
     метал
    1.02
     металла
    0.97
    Activations Density 0.276%

    No Known Activations