© 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. 100491
    Prev
    Next
    INDEX
    Explanations

    various responsibilities, quoted, declared incidents

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

    This neuron fires strongly on the first content words of a document section or heading—i.e. the opening phrase or title of a text block.

    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
     ſeveral
    -1.28
    雠
    -1.21
    कि
    -1.21
    歺
    -1.20
     eigentliche
    -1.18
    鱨
    -1.16
    iiiiiiii
    -1.15
     Emits
    -1.15
    vky
    -1.13
     ojo
    -1.10
    POSITIVE LOGITS
    6
    1.38
    1
    1.24
    7
    1.11
    ➐
    1.11
     trovano
    1.10
     llamó
    1.09
    8
    1.09
     llamaba
    1.07
    ！　
    1.05
     dicono
    1.05
    Activations Density 0.238%

    No Known Activations