© 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. 10981
    Prev
    Next
    INDEX
    Explanations

    situation, family, school

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

    The neuron fires on verbs or short phrases that introduce conditional or explanatory “disclaimer” statements (e.g. “means,” “if …,” “inform,” “contains affiliate links,” “experiences”).

    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
     
    0.86
     Originally
    0.84
    ك
    0.82
    arlo
    0.81
     Efect
    0.80
    ánico
    0.79
     dintre
    0.79
    ofi
    0.78
     Cf
    0.78
    rigley
    0.78
    POSITIVE LOGITS
    𝔰
    0.77
     consoles
    0.76
    ेंगू
    0.75
    和谐
    0.70
     autoComplete
    0.70
    
    0.69
    
    0.69
    ͯ
    0.68
     luxuries
    0.68
    StudioProjects
    0.68
    Activations Density 0.000%

    No Known Activations