© 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. 280
    Prev
    Next
    INDEX
    Explanations

    The neuron is a “frequency” detector that lights up most strongly on high‐frequency English tokens (common function words like “the,” “a,” etc., as well as other very frequent content words).

    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
    '
    0.42
    }
    0.39
    ()
    0.37
     함수
    0.37
    </
    0.36
     ()
    0.35
     and
    0.34
    </th>
    0.34
    </code>
    0.34
    </a>
    0.34
    POSITIVE LOGITS
     proverbial
    0.40
     burgeoning
    0.37
     ਉਹ
    0.36
     smartest
    0.36
     overarching
    0.35
    အတူ
    0.35
     slightest
    0.35
     récit
    0.35
     súa
    0.35
    បែ
    0.35
    Activations Density 5.406%

    No Known Activations