© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2209
    Prev
    Next
    INDEX
    Explanations

    portmanteau or blend of words

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

    The neuron fires on descriptive qualifier words—mainly adjectives and adverbs that categorize or qualify a noun (e.g. “mixed,” “kind,” “Western,” “anglicised,” “usually”).

    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
    ակ
    0.94
     Insulin
    0.91
     Enseñ
    0.83
     Gazi
    0.82
     Fotograf
    0.80
    λύ
    0.79
     Vieira
    0.77
    И
    0.76
     Văn
    0.75
     Patienten
    0.75
    POSITIVE LOGITS
    েলার
    0.77
    tuples
    0.76
     retângulo
    0.76
     chars
    0.75
     start
    0.71
     pumpkins
    0.70
     competitions
    0.70
    就会
    0.69
     considérons
    0.69
     probationary
    0.69
    Activations Density 0.000%

    No Known Activations