© 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. 5662
    Prev
    Next
    INDEX
    Explanations

    wide range

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

    The neuron activates on words that signal a broad range or variety, such as “wide,” “range,” “broad,” and “variety.”

    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
    ์
    0.98
    ная
    0.96
     cuad
    0.94
    ست
    0.93
    нда
    0.92
    ксе
    0.92
    らが
    0.92
    んにちは
    0.91
     furlough
    0.91
     confrontation
    0.90
    POSITIVE LOGITS
     macam
    1.22
     Bets
    1.13
     styles
    1.09
    BUCK
    1.07
    amani
    1.05
    }$,
    1.05
     Styles
    1.04
    ゆる
    1.03
     típus
    1.02
     Sizes
    1.01
    Activations Density 0.154%

    No Known Activations