© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 14296
    Prev
    Next
    INDEX
    Explanations

    standard surf, given silicon, produce quas, crumble there, only to be, its presumed, helping individuals, win the, existing state, clinical practice, before starving

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

    The neuron highlights key content words—topical or informative nouns, adjectives, and technical terms that carry the main subject of a passage.

    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
    댔
    -1.32
    これも
    -1.24
     так
    -1.23
    :“……”
    -1.23
    UTELY
    -1.22
     одним
    -1.22
     одной
    -1.20
     také
    -1.16
    IVAL
    -1.16
     pretty
    -1.15
    POSITIVE LOGITS
     than
    3.23
     as
    1.50
     when
    1.35
     Tugas
    1.21
     because
    1.17
     niż
    1.16
    ⿳
    1.13
     ‘‘
    1.12
     anstatt
    1.12
     coming
    1.10
    Activations Density 0.208%

    No Known Activations