© 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. 11910
    Prev
    Next
    INDEX
    Explanations

    list of options

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

    The neuron selectively activates on major action words—often policy-style verbs—in formal statements.

    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
     carburetor
    0.90
     mainframe
    0.80
     HDTV
    0.77
    所有的
    0.75
     computadora
    0.75
     sinusoid
    0.73
    
    0.73
     automobile
    0.73
     extravag
    0.73
     inexpensive
    0.71
    POSITIVE LOGITS
     Brexit
    0.74
    Migration
    0.71
     carers
    0.70
     liées
    0.69
     Migration
    0.67
    喺
    0.66
    Connectivity
    0.66
    跨境
    0.66
     Connectivity
    0.65
    今回の
    0.64
    Activations Density 0.408%

    No Known Activations