© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112830
    Prev
    Next
    INDEX
    Explanations

    so, such, that

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

    The neuron fires on degree-modifying words (especially “so” and “such”) used for intensification.

    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
    嫋
    -1.54
     finir
    -1.52
    berikut
    -1.48
    遄
    -1.48
     parall
    -1.44
    gemeinschaft
    -1.43
    cortina
    -1.41
    ersatz
    -1.40
     imposible
    -1.39
     obvio
    -1.39
    POSITIVE LOGITS
     they
    1.73
    b
    1.71
     when
    1.63
    e
    1.63
    which
    1.62
    1
    1.61
    p
    1.55
    f
    1.55
    9
    1.47
    \
    1.41
    Activations Density 0.021%

    No Known Activations