© 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. 9847
    Prev
    Next
    INDEX
    Explanations

    goal setting elements

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

    This neuron spots numerical metadata values—especially floating‐point timestamps or similar numeric figures embedded in the text.

    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
    ↵↵
    0.64
    venues
    0.59
    language
    0.54
    visual
    0.54
    必
    0.53
    firestore
    0.52
    content
    0.52
    popular
    0.52
    buds
    0.52
    skyblue
    0.51
    POSITIVE LOGITS
     "/")
    0.55
    विण्यात
    0.55
     Bxf
    0.54
     afgelopen
    0.54
     ойно
    0.53
    वण्यात
    0.53
     '/',
    0.52
    მასრულ
    0.52
     attributing
    0.52
     Nutr
    0.51
    Activations Density 0.007%

    No Known Activations