© 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-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 131042
    Prev
    Next
    INDEX
    Explanations

    huge amounts

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

    The neuron fires on emphatic large-quantity expressions—especially “huge” (and its variants) often paired with “amount (of)”—signaling strong magnifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Biografía
    -1.76
     For
    -1.75
    DESCRIPCIÓN
    -1.73
    涠
    -1.68
    eração
    -1.67
    》,
    -1.64
    ."));
    -1.60
     setiap
    -1.53
    ╽
    -1.52
    。」
    -1.51
    POSITIVE LOGITS
    huge
    1.98
    </h4>
    1.80
     soar
    1.66
     дополнительные
    1.66
     ogrom
    1.63
     huge
    1.59
    greater
    1.58
    Huge
    1.55
    yml
    1.54
     montrer
    1.48
    Activations Density 0.032%

    No Known Activations