© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9101
    Prev
    Next
    INDEX
    Explanations

    The neuron fires most strongly on less-common, more “formal” content words—i.e. relatively rare or advanced-vocabulary tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the word "a" when it precedes technical or descriptive nouns in instructional or informational contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    descriptive modifiers and property-like terms indicating features, settings, or qualitative attributes in technical or informational text.

    oai_token-act-pair · gpt-5Triggered 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
     eukary
    0.75
    ひ
    0.75
     hakk
    0.75
     mü
    0.73
     KON
    0.71
     Aik
    0.71
     fibroblast
    0.70
    撗
    0.70
    Vamos
    0.70
     exfol
    0.70
    POSITIVE LOGITS
    їн
    0.86
    aring
    0.84
    andu
    0.80
    anigans
    0.79
    ama
    0.78
    ining
    0.75
    ৪
    0.75
    etics
    0.75
    osing
    0.73
     યોગ
    0.73
    Activations Density 0.000%

    No Known Activations