© 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. 22-GEMMASCOPE-RES-131K
    7. 13758
    Prev
    Next
    INDEX
    Explanations

    events and developments

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

    The neuron fires strongly on verbs denoting formal presentation or introduction of work or findings (e.g. “presented,” “presenting,” “introduced”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     actuelles
    -1.04
    ueel
    -1.01
    一个
    -1.01
     augus
    -1.00
    αλ
    -0.99
    RRRR
    -0.99
     ét
    -0.98
    forall
    -0.98
    ์
    -0.97
     yves
    -0.95
    POSITIVE LOGITS
     their
    3.16
     themselves
    1.94
    各自
    1.81
     different
    1.74
     свої
    1.70
     különböző
    1.61
     swoich
    1.61
    それぞれ
    1.60
     various
    1.58
     çeşitli
    1.48
    Activations Density 0.034%

    No Known Activations