© 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. 14379
    Prev
    Next
    INDEX
    Explanations

    logo packages; Georgian Dream; zero knowledge protocol; query attempts; single person

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

    The neuron detects emphatic, highly subjective language—particularly exclamatory or strongly evaluative expressions (e.g. “awesome!”, “truly,” “!!!”).

    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
    恆
    0.83
    anvil
    0.83
    știg
    0.82
     št
    0.75
    loo
    0.75
    eep
    0.75
    𝘱
    0.75
    psy
    0.74
     mee
    0.74
    }}>
    0.73
    POSITIVE LOGITS
    きます
    0.76
    が増
    0.73
     cowardly
    0.70
    ا
    0.70
    सबसे
    0.70
    ل
    0.69
    が多く
    0.68
     пишу
    0.67
    iking
    0.66
    骥
    0.66
    Activations Density 0.000%

    No Known Activations