© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4613
    Prev
    Next
    INDEX
    Explanations

    code comments and base cases

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

    The neuron fires strongly on discourse‐cue words—tokens that introduce examples, alternatives, future actions, contrasts, or suppositions (e.g. “not,” “will,” “Examples,” “alternatives,” “suppose”).

    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
     Aircraft
    0.87
     Biographie
    0.77
    컥
    0.76
    `
    0.75
    ?.
    0.75
     constitución
    0.75
    *.
    0.74
     Corporations
    0.73
     Blair
    0.73
     Terrorism
    0.72
    POSITIVE LOGITS
    結果
    0.79
    mensagem
    0.79
     comportement
    0.77
    ル
    0.76
    부
    0.76
    怿
    0.76
    resposta
    0.75
    ਰ
    0.75
     embeddings
    0.74
     közül
    0.73
    Activations Density 0.001%

    No Known Activations