© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 70615
    Prev
    Next
    INDEX
    Explanations

    letting something happen

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

    The neuron fires on dynamic “process” verbs—particularly ones that describe something being allowed or set to proceed over time (e.g. load, run, adsorb, cure, shroud, wither).

    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
    after
    -1.48
     всем
    -1.22
     that
    -1.22
     but
    -1.20
     to
    -1.18
    tiennent
    -1.16
    ening
    -1.11
    也都
    -1.08
     tetrachloride
    -1.04
    慜
    -1.01
    POSITIVE LOGITS
     without
    1.48
    cciale
    1.23
     instead
    1.20
     our
    1.17
     cabang
    1.13
    𝐔
    1.12
     verschillende
    1.11
    endalikan
    1.09
     définiti
    1.09
     infancia
    1.09
    Activations Density 0.114%

    No Known Activations